{"schemaVersion":"agentstaffer.trial-queue.v0","generatedAt":"2026-11-02T01:18:43.849Z","canonicalUrl":"https://agentstaffer.com/trials.json?q=agent-evaluation&sort=trial_readiness","htmlUrl":"https://agentstaffer.com/trials?q=agent-evaluation&sort=trial_readiness","crawlerEntrypoints":{"schemaVersion":"agentstaffer.crawler-entrypoints.v1","sourcePath":"/trials.json","purpose":"Cloudflare shows /trials and /trials.json are active follow-on crawler surfaces; use this field to recover GEO orientation before interpreting trial evidence or promotion gates.","cloudflareSignal":{"observedAt":"2026-06-16T17:08:40.021Z","window":{"start":"2026-06-15T17:08:34.785Z","end":"2026-06-16T17:08:34.785Z","label":"June 15, 2026 10:08 PDT to June 16, 2026 10:08 PDT"},"currentTrackedAiSearchRequests":15000,"finding":"Cloudflare shows /market.json, /compare.json, /api/market-lite, and /trials.json remain active, while /llms.txt and /geo.json had zero tracked AI/search requests in the current 24h window."},"requiredOrientation":[{"path":"/llms.txt","url":"https://agentstaffer.com/llms.txt"},{"path":"/geo.json","url":"https://agentstaffer.com/geo.json"},{"path":"/agent-visibility.json","url":"https://agentstaffer.com/agent-visibility.json"},{"path":"/openai.json","url":"https://agentstaffer.com/openai.json"},{"path":"/ai-crawl-monitor.json","url":"https://agentstaffer.com/ai-crawl-monitor.json"}],"recommendedNextFetches":[{"path":"/promotions.json","url":"https://agentstaffer.com/promotions.json"},{"path":"/compare.json","url":"https://agentstaffer.com/compare.json"},{"path":"/market.json","url":"https://agentstaffer.com/market.json"},{"path":"/agents.json","url":"https://agentstaffer.com/agents.json"},{"path":"/resources.json","url":"https://agentstaffer.com/resources.json"},{"path":"/retrieval.json","url":"https://agentstaffer.com/retrieval.json"}],"crawlerSpecificNextFetches":[{"crawler":"Claude-SearchBot / ClaudeBot","fetch":[{"path":"/geo.json","url":"https://agentstaffer.com/geo.json"},{"path":"/compare.json","url":"https://agentstaffer.com/compare.json"},{"path":"/trials.json","url":"https://agentstaffer.com/trials.json"}]},{"crawler":"OAI-SearchBot","fetch":[{"path":"/openai.json","url":"https://agentstaffer.com/openai.json"},{"path":"/answers.json","url":"https://agentstaffer.com/answers.json"},{"path":"/market.json","url":"https://agentstaffer.com/market.json"},{"path":"/compare.json","url":"https://agentstaffer.com/compare.json"},{"path":"/sitemap.xml","url":"https://agentstaffer.com/sitemap.xml"}]},{"crawler":"ChatGPT-User","fetch":[{"path":"/openai.json","url":"https://agentstaffer.com/openai.json"},{"path":"/llms.txt","url":"https://agentstaffer.com/llms.txt"},{"path":"/answers.json","url":"https://agentstaffer.com/answers.json"},{"path":"/compare.json","url":"https://agentstaffer.com/compare.json"},{"path":"/trials.json","url":"https://agentstaffer.com/trials.json"}]},{"crawler":"PerplexityBot","fetch":[{"path":"/perplexity.json","url":"https://agentstaffer.com/perplexity.json"},{"path":"/geo.json","url":"https://agentstaffer.com/geo.json"},{"path":"/retrieval.json","url":"https://agentstaffer.com/retrieval.json"},{"path":"/compare.json","url":"https://agentstaffer.com/compare.json"}]},{"crawler":"Googlebot / GoogleOther","fetch":[{"path":"/sitemap.xml","url":"https://agentstaffer.com/sitemap.xml"},{"path":"/market.json","url":"https://agentstaffer.com/market.json"},{"path":"/resources.json","url":"https://agentstaffer.com/resources.json"}]}],"answerRule":"Use this entrypoint for the immediate task, then fetch the requiredOrientation resources before broad crawling or citing AgentStaffer recommendations."},"count":18,"trialQueue":[{"id":"npm:awaithumans","slug":"npm-awaithumans","name":"awaithumans","type":"workflow","source":"npm","status":"pending","summary":"HITL infrastructure for AI agents. Your agent calls awaitHuman(), a human reviews via Slack/email/dashboard, agent resumes with a typed response.","profileUrl":"https://agentstaffer.com/mcp/npm-awaithumans","fitScore":61,"roleFamily":"Coding & Development","reviewSignal":"Pending","urls":{"profileJson":"https://agentstaffer.com/mcp/npm-awaithumans/profile.json","installJson":"https://agentstaffer.com/mcp/npm-awaithumans/install.json","trustJson":"https://agentstaffer.com/mcp/npm-awaithumans/trust.json","trialJson":"https://agentstaffer.com/mcp/npm-awaithumans/trial.json"},"queueRank":1,"recommendation":{"label":"Needs review","score":61},"trialPriority":81,"trial":{"title":"Run a small code review trial","input":"Give the agent one low-risk issue or one small pull request and ask for defects, tests, and a patch plan.","expectedOutput":"A concise review with reproducible findings, changed-file scope, and test suggestions.","safetyNotes":["Run in a sandbox or throwaway workspace.","Use non-production data and temporary credentials.","Capture output, logs, install friction, and any permission requests."],"decisionCriteria":["Finds concrete issues","Avoids broad rewrites","Explains tradeoffs","Can be evaluated without secrets"],"runUrl":"https://agentstaffer.com/mcp/npm-awaithumans#trial","runPath":"/mcp/npm-awaithumans#trial","createEndpoint":"https://agentstaffer.com/api/actions/run-trial","createBody":{"listingId":"npm:awaithumans","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}},"launchPacket":{"schemaVersion":"agentstaffer.trial-launch-packet-lite.v1","requiredPermissions":{"schemaVersion":"agentstaffer.trial-requirements.v1","principle":"Use the smallest sandbox permission set needed for the first reversible trial.","requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required","text":"Sandbox workspace: Run the first trial in a throwaway directory or isolated environment."},{"id":"node","label":"Node.js and npm","detail":"Required to run awaithumans.","severity":"required","text":"Node.js and npm: Required to run awaithumans."},{"id":"mcp-client","label":"MCP-capable client","detail":"Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","severity":"required","text":"MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs."},{"id":"device-access","label":"Browser or desktop permissions","detail":"Grant only the minimum UI or device access needed for the first trial.","severity":"conditional","text":"Browser or desktop permissions: Grant only the minimum UI or device access needed for the first trial."},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review","text":"License review: License metadata is missing from the current public profile."}],"humanReadable":["Sandbox workspace: Run the first trial in a throwaway directory or isolated environment.","Node.js and npm: Required to run awaithumans.","MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","Browser or desktop permissions: Grant only the minimum UI or device access needed for the first trial.","License review: License metadata is missing from the current public profile."]},"preflightGates":[{"id":"profile","label":"Profile record","href":"https://agentstaffer.com/mcp/npm-awaithumans/profile.json","passed":true},{"id":"install","label":"Setup guide","href":"https://agentstaffer.com/mcp/npm-awaithumans/install.json","passed":true},{"id":"sandbox","label":"Use sandbox data","passed":true}],"promotionGates":[{"id":"install","label":"Install path reviewed","passed":true},{"id":"trust","label":"Trust evidence reviewed","passed":false},{"id":"trial","label":"Sandbox trial recorded","passed":false}],"stopConditions":["Install asks for production credentials or broad write access.","Source or package cannot be verified from profile evidence."],"createRequest":{"endpoint":"https://agentstaffer.com/api/actions/run-trial","method":"POST","body":{"listingId":"npm:awaithumans","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}}},"safety":{"trustTier":"Needs review","installReadiness":"MCP ready","topRisk":{"level":"medium","label":"No community evaluation","detail":"AgentStaffer users have not submitted a review yet."},"requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required"},{"id":"node","label":"Node.js and npm","detail":"Required to run awaithumans.","severity":"required"},{"id":"mcp-client","label":"MCP-capable client","detail":"Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","severity":"required"},{"id":"device-access","label":"Browser or desktop permissions","detail":"Grant only the minimum UI or device access needed for the first trial.","severity":"conditional"},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review"}],"permissionSummary":{"schemaVersion":"agentstaffer.trial-requirements.v1","principle":"Use the smallest sandbox permission set needed for the first reversible trial.","requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required","text":"Sandbox workspace: Run the first trial in a throwaway directory or isolated environment."},{"id":"node","label":"Node.js and npm","detail":"Required to run awaithumans.","severity":"required","text":"Node.js and npm: Required to run awaithumans."},{"id":"mcp-client","label":"MCP-capable client","detail":"Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","severity":"required","text":"MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs."},{"id":"device-access","label":"Browser or desktop permissions","detail":"Grant only the minimum UI or device access needed for the first trial.","severity":"conditional","text":"Browser or desktop permissions: Grant only the minimum UI or device access needed for the first trial."},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review","text":"License review: License metadata is missing from the current public profile."}],"humanReadable":["Sandbox workspace: Run the first trial in a throwaway directory or isolated environment.","Node.js and npm: Required to run awaithumans.","MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","Browser or desktop permissions: Grant only the minimum UI or device access needed for the first trial.","License review: License metadata is missing from the current public profile."]}}},{"id":"mcpregistry:servers-io-github-hidai25-2fevalview-mcp","slug":"mcpregistry-evalview-mcp","name":"Evalview Mcp","type":"tool","source":"mcpregistry","status":"pending","summary":"Regression testing for AI agents. Golden baselines, CI/CD, LangGraph, CrewAI, OpenAI, Claude.","profileUrl":"https://agentstaffer.com/mcp/mcpregistry-evalview-mcp","fitScore":61,"roleFamily":"Coding & Development","reviewSignal":"Pending","urls":{"profileJson":"https://agentstaffer.com/mcp/mcpregistry-evalview-mcp/profile.json","installJson":"https://agentstaffer.com/mcp/mcpregistry-evalview-mcp/install.json","trustJson":"https://agentstaffer.com/mcp/mcpregistry-evalview-mcp/trust.json","trialJson":"https://agentstaffer.com/mcp/mcpregistry-evalview-mcp/trial.json"},"queueRank":2,"recommendation":{"label":"Needs review","score":61},"trialPriority":81,"trial":{"title":"Run a small code review trial","input":"Give the agent one low-risk issue or one small pull request and ask for defects, tests, and a patch plan.","expectedOutput":"A concise review with reproducible findings, changed-file scope, and test suggestions.","safetyNotes":["Run in a sandbox or throwaway workspace.","Use non-production data and temporary credentials.","Capture output, logs, install friction, and any permission requests."],"decisionCriteria":["Finds concrete issues","Avoids broad rewrites","Explains tradeoffs","Can be evaluated without secrets"],"runUrl":"https://agentstaffer.com/mcp/mcpregistry-evalview-mcp#trial","runPath":"/mcp/mcpregistry-evalview-mcp#trial","createEndpoint":"https://agentstaffer.com/api/actions/run-trial","createBody":{"listingId":"mcpregistry:servers-io-github-hidai25-2fevalview-mcp","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}},"launchPacket":{"schemaVersion":"agentstaffer.trial-launch-packet-lite.v1","requiredPermissions":{"schemaVersion":"agentstaffer.trial-requirements.v1","principle":"Use the smallest sandbox permission set needed for the first reversible trial.","requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required","text":"Sandbox workspace: Run the first trial in a throwaway directory or isolated environment."},{"id":"mcp-client","label":"MCP-capable client","detail":"Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","severity":"required","text":"MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs."},{"id":"terminal","label":"Local terminal","detail":"A CLI path is visible, but the exact command must be verified from source.","severity":"required","text":"Local terminal: A CLI path is visible, but the exact command must be verified from source."},{"id":"temporary-credentials","label":"Temporary credentials","detail":"Use throwaway API keys or scoped test credentials. Do not start with production secrets.","severity":"conditional","text":"Temporary credentials: Use throwaway API keys or scoped test credentials. Do not start with production secrets."},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review","text":"License review: License metadata is missing from the current public profile."}],"humanReadable":["Sandbox workspace: Run the first trial in a throwaway directory or isolated environment.","MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","Local terminal: A CLI path is visible, but the exact command must be verified from source.","Temporary credentials: Use throwaway API keys or scoped test credentials. Do not start with production secrets.","License review: License metadata is missing from the current public profile."]},"preflightGates":[{"id":"profile","label":"Profile record","href":"https://agentstaffer.com/mcp/mcpregistry-evalview-mcp/profile.json","passed":true},{"id":"install","label":"Setup guide","href":"https://agentstaffer.com/mcp/mcpregistry-evalview-mcp/install.json","passed":true},{"id":"sandbox","label":"Use sandbox data","passed":true}],"promotionGates":[{"id":"install","label":"Install path reviewed","passed":true},{"id":"trust","label":"Trust evidence reviewed","passed":false},{"id":"trial","label":"Sandbox trial recorded","passed":false}],"stopConditions":["Install asks for production credentials or broad write access.","Source or package cannot be verified from profile evidence."],"createRequest":{"endpoint":"https://agentstaffer.com/api/actions/run-trial","method":"POST","body":{"listingId":"mcpregistry:servers-io-github-hidai25-2fevalview-mcp","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}}},"safety":{"trustTier":"Needs review","installReadiness":"MCP ready","topRisk":{"level":"medium","label":"No community evaluation","detail":"AgentStaffer users have not submitted a review yet."},"requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required"},{"id":"mcp-client","label":"MCP-capable client","detail":"Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","severity":"required"},{"id":"terminal","label":"Local terminal","detail":"A CLI path is visible, but the exact command must be verified from source.","severity":"required"},{"id":"temporary-credentials","label":"Temporary credentials","detail":"Use throwaway API keys or scoped test credentials. Do not start with production secrets.","severity":"conditional"},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review"}],"permissionSummary":{"schemaVersion":"agentstaffer.trial-requirements.v1","principle":"Use the smallest sandbox permission set needed for the first reversible trial.","requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required","text":"Sandbox workspace: Run the first trial in a throwaway directory or isolated environment."},{"id":"mcp-client","label":"MCP-capable client","detail":"Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","severity":"required","text":"MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs."},{"id":"terminal","label":"Local terminal","detail":"A CLI path is visible, but the exact command must be verified from source.","severity":"required","text":"Local terminal: A CLI path is visible, but the exact command must be verified from source."},{"id":"temporary-credentials","label":"Temporary credentials","detail":"Use throwaway API keys or scoped test credentials. Do not start with production secrets.","severity":"conditional","text":"Temporary credentials: Use throwaway API keys or scoped test credentials. Do not start with production secrets."},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review","text":"License review: License metadata is missing from the current public profile."}],"humanReadable":["Sandbox workspace: Run the first trial in a throwaway directory or isolated environment.","MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","Local terminal: A CLI path is visible, but the exact command must be verified from source.","Temporary credentials: Use throwaway API keys or scoped test credentials. Do not start with production secrets.","License review: License metadata is missing from the current public profile."]}}},{"id":"playbooks:skills-zpankz-mcp-skillset-agent-evaluation","slug":"playbooks-agent-evaluation-skill-by-zpankz-mcp-skillset","name":"agent-evaluation skill by zpankz/mcp-skillset","type":"skill","source":"playbooks","status":"pending","summary":"This skill helps evaluate and improve Claude Code agents by applying multi-dimensional rubrics, non-determinism considerations, and context engineering","profileUrl":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-zpankz-mcp-skillset","fitScore":61,"roleFamily":"Coding & Development","reviewSignal":"Pending","urls":{"profileJson":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-zpankz-mcp-skillset/profile.json","installJson":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-zpankz-mcp-skillset/install.json","trustJson":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-zpankz-mcp-skillset/trust.json","trialJson":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-zpankz-mcp-skillset/trial.json"},"queueRank":3,"recommendation":{"label":"Needs review","score":61},"trialPriority":81,"trial":{"title":"Run a small code review trial","input":"Give the agent one low-risk issue or one small pull request and ask for defects, tests, and a patch plan.","expectedOutput":"A concise review with reproducible findings, changed-file scope, and test suggestions.","safetyNotes":["Run in a sandbox or throwaway workspace.","Use non-production data and temporary credentials.","Capture output, logs, install friction, and any permission requests."],"decisionCriteria":["Finds concrete issues","Avoids broad rewrites","Explains tradeoffs","Can be evaluated without secrets"],"runUrl":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-zpankz-mcp-skillset#trial","runPath":"/skills/playbooks-agent-evaluation-skill-by-zpankz-mcp-skillset#trial","createEndpoint":"https://agentstaffer.com/api/actions/run-trial","createBody":{"listingId":"playbooks:skills-zpankz-mcp-skillset-agent-evaluation","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}},"launchPacket":{"schemaVersion":"agentstaffer.trial-launch-packet-lite.v1","requiredPermissions":{"schemaVersion":"agentstaffer.trial-requirements.v1","principle":"Use the smallest sandbox permission set needed for the first reversible trial.","requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required","text":"Sandbox workspace: Run the first trial in a throwaway directory or isolated environment."},{"id":"mcp-client","label":"MCP-capable client","detail":"Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","severity":"required","text":"MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs."},{"id":"terminal","label":"Local terminal","detail":"A CLI path is visible, but the exact command must be verified from source.","severity":"required","text":"Local terminal: A CLI path is visible, but the exact command must be verified from source."},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review","text":"License review: License metadata is missing from the current public profile."}],"humanReadable":["Sandbox workspace: Run the first trial in a throwaway directory or isolated environment.","MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","Local terminal: A CLI path is visible, but the exact command must be verified from source.","License review: License metadata is missing from the current public profile."]},"preflightGates":[{"id":"profile","label":"Profile record","href":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-zpankz-mcp-skillset/profile.json","passed":true},{"id":"install","label":"Setup guide","href":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-zpankz-mcp-skillset/install.json","passed":true},{"id":"sandbox","label":"Use sandbox data","passed":true}],"promotionGates":[{"id":"install","label":"Install path reviewed","passed":true},{"id":"trust","label":"Trust evidence reviewed","passed":false},{"id":"trial","label":"Sandbox trial recorded","passed":false}],"stopConditions":["Install asks for production credentials or broad write access.","Source or package cannot be verified from profile evidence."],"createRequest":{"endpoint":"https://agentstaffer.com/api/actions/run-trial","method":"POST","body":{"listingId":"playbooks:skills-zpankz-mcp-skillset-agent-evaluation","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}}},"safety":{"trustTier":"Needs review","installReadiness":"MCP ready","topRisk":{"level":"medium","label":"No community evaluation","detail":"AgentStaffer users have not submitted a review yet."},"requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required"},{"id":"mcp-client","label":"MCP-capable client","detail":"Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","severity":"required"},{"id":"terminal","label":"Local terminal","detail":"A CLI path is visible, but the exact command must be verified from source.","severity":"required"},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review"}],"permissionSummary":{"schemaVersion":"agentstaffer.trial-requirements.v1","principle":"Use the smallest sandbox permission set needed for the first reversible trial.","requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required","text":"Sandbox workspace: Run the first trial in a throwaway directory or isolated environment."},{"id":"mcp-client","label":"MCP-capable client","detail":"Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","severity":"required","text":"MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs."},{"id":"terminal","label":"Local terminal","detail":"A CLI path is visible, but the exact command must be verified from source.","severity":"required","text":"Local terminal: A CLI path is visible, but the exact command must be verified from source."},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review","text":"License review: License metadata is missing from the current public profile."}],"humanReadable":["Sandbox workspace: Run the first trial in a throwaway directory or isolated environment.","MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","Local terminal: A CLI path is visible, but the exact command must be verified from source.","License review: License metadata is missing from the current public profile."]}}},{"id":"explainx:skills-neolabhq-context-engineering-kit-customaize-agent-3aagent-evaluation","slug":"explainx-customaize-agent-agent-evaluation","name":"customaize-agent:agent-evaluation","type":"skill","source":"explainx","status":"pending","summary":"Evaluation of agent systems requires different approaches than traditional software or even standard language model applications. Agents make dynamic decisio…","profileUrl":"https://agentstaffer.com/skills/explainx-customaize-agent-agent-evaluation","fitScore":61,"roleFamily":"Coding & Development","reviewSignal":"Pending","urls":{"profileJson":"https://agentstaffer.com/skills/explainx-customaize-agent-agent-evaluation/profile.json","installJson":"https://agentstaffer.com/skills/explainx-customaize-agent-agent-evaluation/install.json","trustJson":"https://agentstaffer.com/skills/explainx-customaize-agent-agent-evaluation/trust.json","trialJson":"https://agentstaffer.com/skills/explainx-customaize-agent-agent-evaluation/trial.json"},"queueRank":4,"recommendation":{"label":"Needs review","score":61},"trialPriority":81,"trial":{"title":"Run a small code review trial","input":"Give the agent one low-risk issue or one small pull request and ask for defects, tests, and a patch plan.","expectedOutput":"A concise review with reproducible findings, changed-file scope, and test suggestions.","safetyNotes":["Run in a sandbox or throwaway workspace.","Use non-production data and temporary credentials.","Capture output, logs, install friction, and any permission requests."],"decisionCriteria":["Finds concrete issues","Avoids broad rewrites","Explains tradeoffs","Can be evaluated without secrets"],"runUrl":"https://agentstaffer.com/skills/explainx-customaize-agent-agent-evaluation#trial","runPath":"/skills/explainx-customaize-agent-agent-evaluation#trial","createEndpoint":"https://agentstaffer.com/api/actions/run-trial","createBody":{"listingId":"explainx:skills-neolabhq-context-engineering-kit-customaize-agent-3aagent-evaluation","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}},"launchPacket":{"schemaVersion":"agentstaffer.trial-launch-packet-lite.v1","requiredPermissions":{"schemaVersion":"agentstaffer.trial-requirements.v1","principle":"Use the smallest sandbox permission set needed for the first reversible trial.","requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required","text":"Sandbox workspace: Run the first trial in a throwaway directory or isolated environment."},{"id":"mcp-client","label":"MCP-capable client","detail":"Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","severity":"required","text":"MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs."},{"id":"terminal","label":"Local terminal","detail":"A CLI path is visible, but the exact command must be verified from source.","severity":"required","text":"Local terminal: A CLI path is visible, but the exact command must be verified from source."},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review","text":"License review: License metadata is missing from the current public profile."}],"humanReadable":["Sandbox workspace: Run the first trial in a throwaway directory or isolated environment.","MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","Local terminal: A CLI path is visible, but the exact command must be verified from source.","License review: License metadata is missing from the current public profile."]},"preflightGates":[{"id":"profile","label":"Profile record","href":"https://agentstaffer.com/skills/explainx-customaize-agent-agent-evaluation/profile.json","passed":true},{"id":"install","label":"Setup guide","href":"https://agentstaffer.com/skills/explainx-customaize-agent-agent-evaluation/install.json","passed":true},{"id":"sandbox","label":"Use sandbox data","passed":true}],"promotionGates":[{"id":"install","label":"Install path reviewed","passed":true},{"id":"trust","label":"Trust evidence reviewed","passed":false},{"id":"trial","label":"Sandbox trial recorded","passed":false}],"stopConditions":["Install asks for production credentials or broad write access.","Source or package cannot be verified from profile evidence."],"createRequest":{"endpoint":"https://agentstaffer.com/api/actions/run-trial","method":"POST","body":{"listingId":"explainx:skills-neolabhq-context-engineering-kit-customaize-agent-3aagent-evaluation","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}}},"safety":{"trustTier":"Needs review","installReadiness":"MCP ready","topRisk":{"level":"medium","label":"No community evaluation","detail":"AgentStaffer users have not submitted a review yet."},"requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required"},{"id":"mcp-client","label":"MCP-capable client","detail":"Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","severity":"required"},{"id":"terminal","label":"Local terminal","detail":"A CLI path is visible, but the exact command must be verified from source.","severity":"required"},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review"}],"permissionSummary":{"schemaVersion":"agentstaffer.trial-requirements.v1","principle":"Use the smallest sandbox permission set needed for the first reversible trial.","requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required","text":"Sandbox workspace: Run the first trial in a throwaway directory or isolated environment."},{"id":"mcp-client","label":"MCP-capable client","detail":"Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","severity":"required","text":"MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs."},{"id":"terminal","label":"Local terminal","detail":"A CLI path is visible, but the exact command must be verified from source.","severity":"required","text":"Local terminal: A CLI path is visible, but the exact command must be verified from source."},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review","text":"License review: License metadata is missing from the current public profile."}],"humanReadable":["Sandbox workspace: Run the first trial in a throwaway directory or isolated environment.","MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","Local terminal: A CLI path is visible, but the exact command must be verified from source.","License review: License metadata is missing from the current public profile."]}}},{"id":"playbooks:skills-eyadsibai-ltk-agent-evaluation","slug":"playbooks-agent-evaluation-skill-by-eyadsibai-ltk","name":"agent-evaluation skill by eyadsibai/ltk","type":"skill","source":"playbooks","status":"pending","summary":"This skill helps you evaluate agent performance, design test sets, and apply multi-dimensional rubrics to improve quality and reliability.","profileUrl":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-eyadsibai-ltk","fitScore":61,"roleFamily":"Coding & Development","reviewSignal":"Pending","urls":{"profileJson":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-eyadsibai-ltk/profile.json","installJson":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-eyadsibai-ltk/install.json","trustJson":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-eyadsibai-ltk/trust.json","trialJson":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-eyadsibai-ltk/trial.json"},"queueRank":5,"recommendation":{"label":"Needs review","score":61},"trialPriority":81,"trial":{"title":"Run a small code review trial","input":"Give the agent one low-risk issue or one small pull request and ask for defects, tests, and a patch plan.","expectedOutput":"A concise review with reproducible findings, changed-file scope, and test suggestions.","safetyNotes":["Run in a sandbox or throwaway workspace.","Use non-production data and temporary credentials.","Capture output, logs, install friction, and any permission requests."],"decisionCriteria":["Finds concrete issues","Avoids broad rewrites","Explains tradeoffs","Can be evaluated without secrets"],"runUrl":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-eyadsibai-ltk#trial","runPath":"/skills/playbooks-agent-evaluation-skill-by-eyadsibai-ltk#trial","createEndpoint":"https://agentstaffer.com/api/actions/run-trial","createBody":{"listingId":"playbooks:skills-eyadsibai-ltk-agent-evaluation","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}},"launchPacket":{"schemaVersion":"agentstaffer.trial-launch-packet-lite.v1","requiredPermissions":{"schemaVersion":"agentstaffer.trial-requirements.v1","principle":"Use the smallest sandbox permission set needed for the first reversible trial.","requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required","text":"Sandbox workspace: Run the first trial in a throwaway directory or isolated environment."},{"id":"mcp-client","label":"MCP-capable client","detail":"Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","severity":"required","text":"MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs."},{"id":"terminal","label":"Local terminal","detail":"A CLI path is visible, but the exact command must be verified from source.","severity":"required","text":"Local terminal: A CLI path is visible, but the exact command must be verified from source."},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review","text":"License review: License metadata is missing from the current public profile."}],"humanReadable":["Sandbox workspace: Run the first trial in a throwaway directory or isolated environment.","MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","Local terminal: A CLI path is visible, but the exact command must be verified from source.","License review: License metadata is missing from the current public profile."]},"preflightGates":[{"id":"profile","label":"Profile record","href":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-eyadsibai-ltk/profile.json","passed":true},{"id":"install","label":"Setup guide","href":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-eyadsibai-ltk/install.json","passed":true},{"id":"sandbox","label":"Use sandbox data","passed":true}],"promotionGates":[{"id":"install","label":"Install path reviewed","passed":true},{"id":"trust","label":"Trust evidence reviewed","passed":false},{"id":"trial","label":"Sandbox trial recorded","passed":false}],"stopConditions":["Install asks for production credentials or broad write access.","Source or package cannot be verified from profile evidence."],"createRequest":{"endpoint":"https://agentstaffer.com/api/actions/run-trial","method":"POST","body":{"listingId":"playbooks:skills-eyadsibai-ltk-agent-evaluation","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}}},"safety":{"trustTier":"Needs review","installReadiness":"MCP ready","topRisk":{"level":"medium","label":"No community evaluation","detail":"AgentStaffer users have not submitted a review yet."},"requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required"},{"id":"mcp-client","label":"MCP-capable client","detail":"Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","severity":"required"},{"id":"terminal","label":"Local terminal","detail":"A CLI path is visible, but the exact command must be verified from source.","severity":"required"},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review"}],"permissionSummary":{"schemaVersion":"agentstaffer.trial-requirements.v1","principle":"Use the smallest sandbox permission set needed for the first reversible trial.","requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required","text":"Sandbox workspace: Run the first trial in a throwaway directory or isolated environment."},{"id":"mcp-client","label":"MCP-capable client","detail":"Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","severity":"required","text":"MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs."},{"id":"terminal","label":"Local terminal","detail":"A CLI path is visible, but the exact command must be verified from source.","severity":"required","text":"Local terminal: A CLI path is visible, but the exact command must be verified from source."},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review","text":"License review: License metadata is missing from the current public profile."}],"humanReadable":["Sandbox workspace: Run the first trial in a throwaway directory or isolated environment.","MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","Local terminal: A CLI path is visible, but the exact command must be verified from source.","License review: License metadata is missing from the current public profile."]}}},{"id":"playbooks:skills-supercent-io-skills-template-agent-evaluation","slug":"playbooks-agent-evaluation-skill-by-supercent-io-skills-template","name":"agent-evaluation skill by supercent-io/skills-template","type":"skill","source":"playbooks","status":"pending","summary":"This skill helps you design, implement, and maintain comprehensive AI agent evaluation systems across coding, conversational, and research domains.","profileUrl":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-supercent-io-skills-template","fitScore":61,"roleFamily":"Coding & Development","reviewSignal":"Pending","urls":{"profileJson":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-supercent-io-skills-template/profile.json","installJson":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-supercent-io-skills-template/install.json","trustJson":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-supercent-io-skills-template/trust.json","trialJson":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-supercent-io-skills-template/trial.json"},"queueRank":6,"recommendation":{"label":"Needs review","score":61},"trialPriority":81,"trial":{"title":"Run a small code review trial","input":"Give the agent one low-risk issue or one small pull request and ask for defects, tests, and a patch plan.","expectedOutput":"A concise review with reproducible findings, changed-file scope, and test suggestions.","safetyNotes":["Run in a sandbox or throwaway workspace.","Use non-production data and temporary credentials.","Capture output, logs, install friction, and any permission requests."],"decisionCriteria":["Finds concrete issues","Avoids broad rewrites","Explains tradeoffs","Can be evaluated without secrets"],"runUrl":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-supercent-io-skills-template#trial","runPath":"/skills/playbooks-agent-evaluation-skill-by-supercent-io-skills-template#trial","createEndpoint":"https://agentstaffer.com/api/actions/run-trial","createBody":{"listingId":"playbooks:skills-supercent-io-skills-template-agent-evaluation","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}},"launchPacket":{"schemaVersion":"agentstaffer.trial-launch-packet-lite.v1","requiredPermissions":{"schemaVersion":"agentstaffer.trial-requirements.v1","principle":"Use the smallest sandbox permission set needed for the first reversible trial.","requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required","text":"Sandbox workspace: Run the first trial in a throwaway directory or isolated environment."},{"id":"mcp-client","label":"MCP-capable client","detail":"Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","severity":"required","text":"MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs."},{"id":"terminal","label":"Local terminal","detail":"A CLI path is visible, but the exact command must be verified from source.","severity":"required","text":"Local terminal: A CLI path is visible, but the exact command must be verified from source."},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review","text":"License review: License metadata is missing from the current public profile."}],"humanReadable":["Sandbox workspace: Run the first trial in a throwaway directory or isolated environment.","MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","Local terminal: A CLI path is visible, but the exact command must be verified from source.","License review: License metadata is missing from the current public profile."]},"preflightGates":[{"id":"profile","label":"Profile record","href":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-supercent-io-skills-template/profile.json","passed":true},{"id":"install","label":"Setup guide","href":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-supercent-io-skills-template/install.json","passed":true},{"id":"sandbox","label":"Use sandbox data","passed":true}],"promotionGates":[{"id":"install","label":"Install path reviewed","passed":true},{"id":"trust","label":"Trust evidence reviewed","passed":false},{"id":"trial","label":"Sandbox trial recorded","passed":false}],"stopConditions":["Install asks for production credentials or broad write access.","Source or package cannot be verified from profile evidence."],"createRequest":{"endpoint":"https://agentstaffer.com/api/actions/run-trial","method":"POST","body":{"listingId":"playbooks:skills-supercent-io-skills-template-agent-evaluation","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}}},"safety":{"trustTier":"Needs review","installReadiness":"MCP ready","topRisk":{"level":"medium","label":"No community evaluation","detail":"AgentStaffer users have not submitted a review yet."},"requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required"},{"id":"mcp-client","label":"MCP-capable client","detail":"Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","severity":"required"},{"id":"terminal","label":"Local terminal","detail":"A CLI path is visible, but the exact command must be verified from source.","severity":"required"},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review"}],"permissionSummary":{"schemaVersion":"agentstaffer.trial-requirements.v1","principle":"Use the smallest sandbox permission set needed for the first reversible trial.","requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required","text":"Sandbox workspace: Run the first trial in a throwaway directory or isolated environment."},{"id":"mcp-client","label":"MCP-capable client","detail":"Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","severity":"required","text":"MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs."},{"id":"terminal","label":"Local terminal","detail":"A CLI path is visible, but the exact command must be verified from source.","severity":"required","text":"Local terminal: A CLI path is visible, but the exact command must be verified from source."},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review","text":"License review: License metadata is missing from the current public profile."}],"humanReadable":["Sandbox workspace: Run the first trial in a throwaway directory or isolated environment.","MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","Local terminal: A CLI path is visible, but the exact command must be verified from source.","License review: License metadata is missing from the current public profile."]}}},{"id":"playbooks:skills-omer-metin-skills-for-antigravity-agent-evaluation","slug":"playbooks-agent-evaluation-skill-by-omer-metin-skills-for-antigravity","name":"agent-evaluation skill by omer-metin/skills-for-antigravity","type":"skill","source":"playbooks","status":"pending","summary":"This skill helps you rigorously evaluate LLM agents using behavioral contracts, adversarial testing, and production monitoring to predict real-world","profileUrl":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-omer-metin-skills-for-antigravity","fitScore":55,"roleFamily":"Workflow Automation","reviewSignal":"Pending","urls":{"profileJson":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-omer-metin-skills-for-antigravity/profile.json","installJson":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-omer-metin-skills-for-antigravity/install.json","trustJson":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-omer-metin-skills-for-antigravity/trust.json","trialJson":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-omer-metin-skills-for-antigravity/trial.json"},"queueRank":7,"recommendation":{"label":"Needs review","score":55},"trialPriority":75,"trial":{"title":"Run a dry-run workflow trial","input":"Connect the workflow to mock inputs and ask it to produce a trace without touching production systems.","expectedOutput":"A step-by-step run log with inputs, outputs, failures, and rollback notes.","safetyNotes":["Run in a sandbox or throwaway workspace.","Use non-production data and temporary credentials.","Capture output, logs, install friction, and any permission requests."],"decisionCriteria":["Shows observable steps","Handles failure paths","Documents required credentials","Can run without side effects"],"runUrl":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-omer-metin-skills-for-antigravity#trial","runPath":"/skills/playbooks-agent-evaluation-skill-by-omer-metin-skills-for-antigravity#trial","createEndpoint":"https://agentstaffer.com/api/actions/run-trial","createBody":{"listingId":"playbooks:skills-omer-metin-skills-for-antigravity-agent-evaluation","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}},"launchPacket":{"schemaVersion":"agentstaffer.trial-launch-packet-lite.v1","requiredPermissions":{"schemaVersion":"agentstaffer.trial-requirements.v1","principle":"Use the smallest sandbox permission set needed for the first reversible trial.","requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required","text":"Sandbox workspace: Run the first trial in a throwaway directory or isolated environment."},{"id":"mcp-client","label":"MCP-capable client","detail":"Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","severity":"required","text":"MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs."},{"id":"terminal","label":"Local terminal","detail":"A CLI path is visible, but the exact command must be verified from source.","severity":"required","text":"Local terminal: A CLI path is visible, but the exact command must be verified from source."},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review","text":"License review: License metadata is missing from the current public profile."}],"humanReadable":["Sandbox workspace: Run the first trial in a throwaway directory or isolated environment.","MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","Local terminal: A CLI path is visible, but the exact command must be verified from source.","License review: License metadata is missing from the current public profile."]},"preflightGates":[{"id":"profile","label":"Profile record","href":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-omer-metin-skills-for-antigravity/profile.json","passed":true},{"id":"install","label":"Setup guide","href":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-omer-metin-skills-for-antigravity/install.json","passed":true},{"id":"sandbox","label":"Use sandbox data","passed":true}],"promotionGates":[{"id":"install","label":"Install path reviewed","passed":true},{"id":"trust","label":"Trust evidence reviewed","passed":false},{"id":"trial","label":"Sandbox trial recorded","passed":false}],"stopConditions":["Install asks for production credentials or broad write access.","Source or package cannot be verified from profile evidence."],"createRequest":{"endpoint":"https://agentstaffer.com/api/actions/run-trial","method":"POST","body":{"listingId":"playbooks:skills-omer-metin-skills-for-antigravity-agent-evaluation","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}}},"safety":{"trustTier":"Needs review","installReadiness":"MCP ready","topRisk":{"level":"medium","label":"No community evaluation","detail":"AgentStaffer users have not submitted a review yet."},"requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required"},{"id":"mcp-client","label":"MCP-capable client","detail":"Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","severity":"required"},{"id":"terminal","label":"Local terminal","detail":"A CLI path is visible, but the exact command must be verified from source.","severity":"required"},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review"}],"permissionSummary":{"schemaVersion":"agentstaffer.trial-requirements.v1","principle":"Use the smallest sandbox permission set needed for the first reversible trial.","requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required","text":"Sandbox workspace: Run the first trial in a throwaway directory or isolated environment."},{"id":"mcp-client","label":"MCP-capable client","detail":"Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","severity":"required","text":"MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs."},{"id":"terminal","label":"Local terminal","detail":"A CLI path is visible, but the exact command must be verified from source.","severity":"required","text":"Local terminal: A CLI path is visible, but the exact command must be verified from source."},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review","text":"License review: License metadata is missing from the current public profile."}],"humanReadable":["Sandbox workspace: Run the first trial in a throwaway directory or isolated environment.","MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","Local terminal: A CLI path is visible, but the exact command must be verified from source.","License review: License metadata is missing from the current public profile."]}}},{"id":"playbooks:skills-openclaw-skills-agent-evaluation","slug":"playbooks-agent-evaluation-skill-by-openclaw-skills","name":"agent-evaluation skill by openclaw/skills","type":"skill","source":"playbooks","status":"pending","summary":"This skill helps you design and run rigorous agent evaluations, balancing benchmark results with real-world reliability to prevent production failures.","profileUrl":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-openclaw-skills","fitScore":55,"roleFamily":"Workflow Automation","reviewSignal":"Pending","urls":{"profileJson":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-openclaw-skills/profile.json","installJson":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-openclaw-skills/install.json","trustJson":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-openclaw-skills/trust.json","trialJson":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-openclaw-skills/trial.json"},"queueRank":8,"recommendation":{"label":"Needs review","score":55},"trialPriority":75,"trial":{"title":"Run a dry-run workflow trial","input":"Connect the workflow to mock inputs and ask it to produce a trace without touching production systems.","expectedOutput":"A step-by-step run log with inputs, outputs, failures, and rollback notes.","safetyNotes":["Run in a sandbox or throwaway workspace.","Use non-production data and temporary credentials.","Capture output, logs, install friction, and any permission requests."],"decisionCriteria":["Shows observable steps","Handles failure paths","Documents required credentials","Can run without side effects"],"runUrl":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-openclaw-skills#trial","runPath":"/skills/playbooks-agent-evaluation-skill-by-openclaw-skills#trial","createEndpoint":"https://agentstaffer.com/api/actions/run-trial","createBody":{"listingId":"playbooks:skills-openclaw-skills-agent-evaluation","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}},"launchPacket":{"schemaVersion":"agentstaffer.trial-launch-packet-lite.v1","requiredPermissions":{"schemaVersion":"agentstaffer.trial-requirements.v1","principle":"Use the smallest sandbox permission set needed for the first reversible trial.","requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required","text":"Sandbox workspace: Run the first trial in a throwaway directory or isolated environment."},{"id":"mcp-client","label":"MCP-capable client","detail":"Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","severity":"required","text":"MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs."},{"id":"terminal","label":"Local terminal","detail":"A CLI path is visible, but the exact command must be verified from source.","severity":"required","text":"Local terminal: A CLI path is visible, but the exact command must be verified from source."},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review","text":"License review: License metadata is missing from the current public profile."}],"humanReadable":["Sandbox workspace: Run the first trial in a throwaway directory or isolated environment.","MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","Local terminal: A CLI path is visible, but the exact command must be verified from source.","License review: License metadata is missing from the current public profile."]},"preflightGates":[{"id":"profile","label":"Profile record","href":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-openclaw-skills/profile.json","passed":true},{"id":"install","label":"Setup guide","href":"https://agentstaffer.com/skills/playbooks-agent-evaluation-skill-by-openclaw-skills/install.json","passed":true},{"id":"sandbox","label":"Use sandbox data","passed":true}],"promotionGates":[{"id":"install","label":"Install path reviewed","passed":true},{"id":"trust","label":"Trust evidence reviewed","passed":false},{"id":"trial","label":"Sandbox trial recorded","passed":false}],"stopConditions":["Install asks for production credentials or broad write access.","Source or package cannot be verified from profile evidence."],"createRequest":{"endpoint":"https://agentstaffer.com/api/actions/run-trial","method":"POST","body":{"listingId":"playbooks:skills-openclaw-skills-agent-evaluation","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}}},"safety":{"trustTier":"Needs review","installReadiness":"MCP ready","topRisk":{"level":"medium","label":"No community evaluation","detail":"AgentStaffer users have not submitted a review yet."},"requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required"},{"id":"mcp-client","label":"MCP-capable client","detail":"Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","severity":"required"},{"id":"terminal","label":"Local terminal","detail":"A CLI path is visible, but the exact command must be verified from source.","severity":"required"},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review"}],"permissionSummary":{"schemaVersion":"agentstaffer.trial-requirements.v1","principle":"Use the smallest sandbox permission set needed for the first reversible trial.","requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required","text":"Sandbox workspace: Run the first trial in a throwaway directory or isolated environment."},{"id":"mcp-client","label":"MCP-capable client","detail":"Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","severity":"required","text":"MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs."},{"id":"terminal","label":"Local terminal","detail":"A CLI path is visible, but the exact command must be verified from source.","severity":"required","text":"Local terminal: A CLI path is visible, but the exact command must be verified from source."},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review","text":"License review: License metadata is missing from the current public profile."}],"humanReadable":["Sandbox workspace: Run the first trial in a throwaway directory or isolated environment.","MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","Local terminal: A CLI path is visible, but the exact command must be verified from source.","License review: License metadata is missing from the current public profile."]}}},{"id":"mcpservers:servers-iris-eval-mcp-server","slug":"mcpservers-iris-mcp-server","name":"Iris MCP Server","type":"tool","source":"mcpservers","status":"pending","summary":"MCP-native agent evaluation and observability server with trace logging, output quality evaluation, cost tracking, 12 built-in eval rules, real-time...","profileUrl":"https://agentstaffer.com/mcp/mcpservers-iris-mcp-server","fitScore":61,"roleFamily":"Coding & Development","reviewSignal":"Pending","urls":{"profileJson":"https://agentstaffer.com/mcp/mcpservers-iris-mcp-server/profile.json","installJson":"https://agentstaffer.com/mcp/mcpservers-iris-mcp-server/install.json","trustJson":"https://agentstaffer.com/mcp/mcpservers-iris-mcp-server/trust.json","trialJson":"https://agentstaffer.com/mcp/mcpservers-iris-mcp-server/trial.json"},"queueRank":9,"recommendation":{"label":"Needs review","score":61},"trialPriority":81,"trial":{"title":"Run a small code review trial","input":"Give the agent one low-risk issue or one small pull request and ask for defects, tests, and a patch plan.","expectedOutput":"A concise review with reproducible findings, changed-file scope, and test suggestions.","safetyNotes":["Run in a sandbox or throwaway workspace.","Use non-production data and temporary credentials.","Capture output, logs, install friction, and any permission requests."],"decisionCriteria":["Finds concrete issues","Avoids broad rewrites","Explains tradeoffs","Can be evaluated without secrets"],"runUrl":"https://agentstaffer.com/mcp/mcpservers-iris-mcp-server#trial","runPath":"/mcp/mcpservers-iris-mcp-server#trial","createEndpoint":"https://agentstaffer.com/api/actions/run-trial","createBody":{"listingId":"mcpservers:servers-iris-eval-mcp-server","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}},"launchPacket":{"schemaVersion":"agentstaffer.trial-launch-packet-lite.v1","requiredPermissions":{"schemaVersion":"agentstaffer.trial-requirements.v1","principle":"Use the smallest sandbox permission set needed for the first reversible trial.","requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required","text":"Sandbox workspace: Run the first trial in a throwaway directory or isolated environment."},{"id":"mcp-client","label":"MCP-capable client","detail":"Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","severity":"required","text":"MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs."},{"id":"terminal","label":"Local terminal","detail":"A CLI path is visible, but the exact command must be verified from source.","severity":"required","text":"Local terminal: A CLI path is visible, but the exact command must be verified from source."},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review","text":"License review: License metadata is missing from the current public profile."}],"humanReadable":["Sandbox workspace: Run the first trial in a throwaway directory or isolated environment.","MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","Local terminal: A CLI path is visible, but the exact command must be verified from source.","License review: License metadata is missing from the current public profile."]},"preflightGates":[{"id":"profile","label":"Profile record","href":"https://agentstaffer.com/mcp/mcpservers-iris-mcp-server/profile.json","passed":true},{"id":"install","label":"Setup guide","href":"https://agentstaffer.com/mcp/mcpservers-iris-mcp-server/install.json","passed":true},{"id":"sandbox","label":"Use sandbox data","passed":true}],"promotionGates":[{"id":"install","label":"Install path reviewed","passed":true},{"id":"trust","label":"Trust evidence reviewed","passed":false},{"id":"trial","label":"Sandbox trial recorded","passed":false}],"stopConditions":["Install asks for production credentials or broad write access.","Source or package cannot be verified from profile evidence."],"createRequest":{"endpoint":"https://agentstaffer.com/api/actions/run-trial","method":"POST","body":{"listingId":"mcpservers:servers-iris-eval-mcp-server","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}}},"safety":{"trustTier":"Needs review","installReadiness":"MCP ready","topRisk":{"level":"medium","label":"No community evaluation","detail":"AgentStaffer users have not submitted a review yet."},"requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required"},{"id":"mcp-client","label":"MCP-capable client","detail":"Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","severity":"required"},{"id":"terminal","label":"Local terminal","detail":"A CLI path is visible, but the exact command must be verified from source.","severity":"required"},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review"}],"permissionSummary":{"schemaVersion":"agentstaffer.trial-requirements.v1","principle":"Use the smallest sandbox permission set needed for the first reversible trial.","requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required","text":"Sandbox workspace: Run the first trial in a throwaway directory or isolated environment."},{"id":"mcp-client","label":"MCP-capable client","detail":"Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","severity":"required","text":"MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs."},{"id":"terminal","label":"Local terminal","detail":"A CLI path is visible, but the exact command must be verified from source.","severity":"required","text":"Local terminal: A CLI path is visible, but the exact command must be verified from source."},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review","text":"License review: License metadata is missing from the current public profile."}],"humanReadable":["Sandbox workspace: Run the first trial in a throwaway directory or isolated environment.","MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","Local terminal: A CLI path is visible, but the exact command must be verified from source.","License review: License metadata is missing from the current public profile."]}}},{"id":"claudemarketplaces:skills-supercent-io-skills-template-agent-evaluation","slug":"claudemarketplaces-agent-evaluation","name":"agent-evaluation","type":"skill","source":"claudemarketplaces","status":"pending","summary":"A structured approach to evaluating AI agents that covers the full lifecycle from initial design to production monitoring. You get an 8-step roadmap starting with task creation and moving through environment isolation, grading strategies, and saturation detection. It breaks down three grader types (code-based, model-based, human) with practical trade-offs f...","profileUrl":"https://agentstaffer.com/skills/claudemarketplaces-agent-evaluation","fitScore":48,"roleFamily":"Coding & Development","reviewSignal":"Pending","urls":{"profileJson":"https://agentstaffer.com/skills/claudemarketplaces-agent-evaluation/profile.json","installJson":"https://agentstaffer.com/skills/claudemarketplaces-agent-evaluation/install.json","trustJson":"https://agentstaffer.com/skills/claudemarketplaces-agent-evaluation/trust.json","trialJson":"https://agentstaffer.com/skills/claudemarketplaces-agent-evaluation/trial.json"},"queueRank":10,"recommendation":{"label":"Needs review","score":48},"trialPriority":68,"trial":{"title":"Run a small code review trial","input":"Give the agent one low-risk issue or one small pull request and ask for defects, tests, and a patch plan.","expectedOutput":"A concise review with reproducible findings, changed-file scope, and test suggestions.","safetyNotes":["Run in a sandbox or throwaway workspace.","Use non-production data and temporary credentials.","Capture output, logs, install friction, and any permission requests."],"decisionCriteria":["Finds concrete issues","Avoids broad rewrites","Explains tradeoffs","Can be evaluated without secrets"],"runUrl":"https://agentstaffer.com/skills/claudemarketplaces-agent-evaluation#trial","runPath":"/skills/claudemarketplaces-agent-evaluation#trial","createEndpoint":"https://agentstaffer.com/api/actions/run-trial","createBody":{"listingId":"claudemarketplaces:skills-supercent-io-skills-template-agent-evaluation","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}},"launchPacket":{"schemaVersion":"agentstaffer.trial-launch-packet-lite.v1","requiredPermissions":{"schemaVersion":"agentstaffer.trial-requirements.v1","principle":"Use the smallest sandbox permission set needed for the first reversible trial.","requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required","text":"Sandbox workspace: Run the first trial in a throwaway directory or isolated environment."},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review","text":"License review: License metadata is missing from the current public profile."}],"humanReadable":["Sandbox workspace: Run the first trial in a throwaway directory or isolated environment.","License review: License metadata is missing from the current public profile."]},"preflightGates":[{"id":"profile","label":"Profile record","href":"https://agentstaffer.com/skills/claudemarketplaces-agent-evaluation/profile.json","passed":true},{"id":"install","label":"Setup guide","href":"https://agentstaffer.com/skills/claudemarketplaces-agent-evaluation/install.json","passed":true},{"id":"sandbox","label":"Use sandbox data","passed":true}],"promotionGates":[{"id":"install","label":"Install path reviewed","passed":true},{"id":"trust","label":"Trust evidence reviewed","passed":false},{"id":"trial","label":"Sandbox trial recorded","passed":false}],"stopConditions":["Install asks for production credentials or broad write access.","Source or package cannot be verified from profile evidence."],"createRequest":{"endpoint":"https://agentstaffer.com/api/actions/run-trial","method":"POST","body":{"listingId":"claudemarketplaces:skills-supercent-io-skills-template-agent-evaluation","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}}},"safety":{"trustTier":"Needs review","installReadiness":"Source guide","topRisk":{"level":"medium","label":"No community evaluation","detail":"AgentStaffer users have not submitted a review yet."},"requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required"},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review"}],"permissionSummary":{"schemaVersion":"agentstaffer.trial-requirements.v1","principle":"Use the smallest sandbox permission set needed for the first reversible trial.","requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required","text":"Sandbox workspace: Run the first trial in a throwaway directory or isolated environment."},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review","text":"License review: License metadata is missing from the current public profile."}],"humanReadable":["Sandbox workspace: Run the first trial in a throwaway directory or isolated environment.","License review: License metadata is missing from the current public profile."]}}},{"id":"remoteopenclaw:skills-neolabhq-context-engineering-kit-agent-evaluation","slug":"remoteopenclaw-agent-evaluation","name":"agent-evaluation","type":"skill","source":"remoteopenclaw","status":"pending","summary":"Evaluate and improve Claude Code commands, skills, and agents. Use when testing prompt effectiveness, validating context engineering choices, or measuring improvement quality.","profileUrl":"https://agentstaffer.com/skills/remoteopenclaw-agent-evaluation","fitScore":42,"roleFamily":"Coding & Development","reviewSignal":"Pending","urls":{"profileJson":"https://agentstaffer.com/skills/remoteopenclaw-agent-evaluation/profile.json","installJson":"https://agentstaffer.com/skills/remoteopenclaw-agent-evaluation/install.json","trustJson":"https://agentstaffer.com/skills/remoteopenclaw-agent-evaluation/trust.json","trialJson":"https://agentstaffer.com/skills/remoteopenclaw-agent-evaluation/trial.json"},"queueRank":11,"recommendation":{"label":"Needs review","score":42},"trialPriority":62,"trial":{"title":"Run a small code review trial","input":"Give the agent one low-risk issue or one small pull request and ask for defects, tests, and a patch plan.","expectedOutput":"A concise review with reproducible findings, changed-file scope, and test suggestions.","safetyNotes":["Run in a sandbox or throwaway workspace.","Use non-production data and temporary credentials.","Capture output, logs, install friction, and any permission requests."],"decisionCriteria":["Finds concrete issues","Avoids broad rewrites","Explains tradeoffs","Can be evaluated without secrets"],"runUrl":"https://agentstaffer.com/skills/remoteopenclaw-agent-evaluation#trial","runPath":"/skills/remoteopenclaw-agent-evaluation#trial","createEndpoint":"https://agentstaffer.com/api/actions/run-trial","createBody":{"listingId":"remoteopenclaw:skills-neolabhq-context-engineering-kit-agent-evaluation","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}},"launchPacket":{"schemaVersion":"agentstaffer.trial-launch-packet-lite.v1","requiredPermissions":{"schemaVersion":"agentstaffer.trial-requirements.v1","principle":"Use the smallest sandbox permission set needed for the first reversible trial.","requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required","text":"Sandbox workspace: Run the first trial in a throwaway directory or isolated environment."},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review","text":"License review: License metadata is missing from the current public profile."}],"humanReadable":["Sandbox workspace: Run the first trial in a throwaway directory or isolated environment.","License review: License metadata is missing from the current public profile."]},"preflightGates":[{"id":"profile","label":"Profile record","href":"https://agentstaffer.com/skills/remoteopenclaw-agent-evaluation/profile.json","passed":true},{"id":"install","label":"Setup guide","href":"https://agentstaffer.com/skills/remoteopenclaw-agent-evaluation/install.json","passed":true},{"id":"sandbox","label":"Use sandbox data","passed":true}],"promotionGates":[{"id":"install","label":"Install path reviewed","passed":true},{"id":"trust","label":"Trust evidence reviewed","passed":false},{"id":"trial","label":"Sandbox trial recorded","passed":false}],"stopConditions":["Install asks for production credentials or broad write access.","Source or package cannot be verified from profile evidence."],"createRequest":{"endpoint":"https://agentstaffer.com/api/actions/run-trial","method":"POST","body":{"listingId":"remoteopenclaw:skills-neolabhq-context-engineering-kit-agent-evaluation","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}}},"safety":{"trustTier":"Needs review","installReadiness":"Source guide","topRisk":{"level":"medium","label":"No community evaluation","detail":"AgentStaffer users have not submitted a review yet."},"requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required"},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review"}],"permissionSummary":{"schemaVersion":"agentstaffer.trial-requirements.v1","principle":"Use the smallest sandbox permission set needed for the first reversible trial.","requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required","text":"Sandbox workspace: Run the first trial in a throwaway directory or isolated environment."},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review","text":"License review: License metadata is missing from the current public profile."}],"humanReadable":["Sandbox workspace: Run the first trial in a throwaway directory or isolated environment.","License review: License metadata is missing from the current public profile."]}}},{"id":"npm:agent-reliability","slug":"npm-agent-reliability","name":"agent-reliability","type":"workflow","source":"npm","status":"pending","summary":"End-to-end reliability testing for multi-step AI agent workflows. Tests what DeepEval and Langfuse can't: cascading failures across 10+ step chains.","profileUrl":"https://agentstaffer.com/agents/npm-agent-reliability","fitScore":38,"roleFamily":"Coding & Development","reviewSignal":"Pending","urls":{"profileJson":"https://agentstaffer.com/agents/npm-agent-reliability/profile.json","installJson":"https://agentstaffer.com/agents/npm-agent-reliability/install.json","trustJson":"https://agentstaffer.com/agents/npm-agent-reliability/trust.json","trialJson":"https://agentstaffer.com/agents/npm-agent-reliability/trial.json"},"queueRank":12,"recommendation":{"label":"Needs review","score":38},"trialPriority":58,"trial":{"title":"Run a small code review trial","input":"Give the agent one low-risk issue or one small pull request and ask for defects, tests, and a patch plan.","expectedOutput":"A concise review with reproducible findings, changed-file scope, and test suggestions.","safetyNotes":["Run in a sandbox or throwaway workspace.","Use non-production data and temporary credentials.","Capture output, logs, install friction, and any permission requests."],"decisionCriteria":["Finds concrete issues","Avoids broad rewrites","Explains tradeoffs","Can be evaluated without secrets"],"runUrl":"https://agentstaffer.com/agents/npm-agent-reliability#trial","runPath":"/agents/npm-agent-reliability#trial","createEndpoint":"https://agentstaffer.com/api/actions/run-trial","createBody":{"listingId":"npm:agent-reliability","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}},"launchPacket":{"schemaVersion":"agentstaffer.trial-launch-packet-lite.v1","requiredPermissions":{"schemaVersion":"agentstaffer.trial-requirements.v1","principle":"Use the smallest sandbox permission set needed for the first reversible trial.","requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required","text":"Sandbox workspace: Run the first trial in a throwaway directory or isolated environment."},{"id":"node","label":"Node.js and npm","detail":"Required to run agent-reliability.","severity":"required","text":"Node.js and npm: Required to run agent-reliability."},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review","text":"License review: License metadata is missing from the current public profile."}],"humanReadable":["Sandbox workspace: Run the first trial in a throwaway directory or isolated environment.","Node.js and npm: Required to run agent-reliability.","License review: License metadata is missing from the current public profile."]},"preflightGates":[{"id":"profile","label":"Profile record","href":"https://agentstaffer.com/agents/npm-agent-reliability/profile.json","passed":true},{"id":"install","label":"Setup guide","href":"https://agentstaffer.com/agents/npm-agent-reliability/install.json","passed":true},{"id":"sandbox","label":"Use sandbox data","passed":true}],"promotionGates":[{"id":"install","label":"Install path reviewed","passed":true},{"id":"trust","label":"Trust evidence reviewed","passed":false},{"id":"trial","label":"Sandbox trial recorded","passed":false}],"stopConditions":["Install asks for production credentials or broad write access.","Source or package cannot be verified from profile evidence."],"createRequest":{"endpoint":"https://agentstaffer.com/api/actions/run-trial","method":"POST","body":{"listingId":"npm:agent-reliability","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}}},"safety":{"trustTier":"Needs review","installReadiness":"CLI ready","topRisk":{"level":"medium","label":"No community evaluation","detail":"AgentStaffer users have not submitted a review yet."},"requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required"},{"id":"node","label":"Node.js and npm","detail":"Required to run agent-reliability.","severity":"required"},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review"}],"permissionSummary":{"schemaVersion":"agentstaffer.trial-requirements.v1","principle":"Use the smallest sandbox permission set needed for the first reversible trial.","requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required","text":"Sandbox workspace: Run the first trial in a throwaway directory or isolated environment."},{"id":"node","label":"Node.js and npm","detail":"Required to run agent-reliability.","severity":"required","text":"Node.js and npm: Required to run agent-reliability."},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review","text":"License review: License metadata is missing from the current public profile."}],"humanReadable":["Sandbox workspace: Run the first trial in a throwaway directory or isolated environment.","Node.js and npm: Required to run agent-reliability.","License review: License metadata is missing from the current public profile."]}}},{"id":"agensi:skills-benchmarking-ai-agents-beyond-models","slug":"agensi-benchmarking-ai-agents-beyond-models","name":"Benchmarking ai Agents Beyond Models","type":"skill","source":"agensi","status":"pending","summary":"Published AI benchmarks measure brains in jars. They test models in isolation or within a single reference harness — and then attribute all performance to the model. This skill teaches you to decompose agent performance into its two actual components: model capability and harness multiplier. The result is evaluations that predict real-world behavior instead...","profileUrl":"https://agentstaffer.com/skills/agensi-benchmarking-ai-agents-beyond-models","fitScore":37,"roleFamily":"Coding & Development","reviewSignal":"Pending","urls":{"profileJson":"https://agentstaffer.com/skills/agensi-benchmarking-ai-agents-beyond-models/profile.json","installJson":"https://agentstaffer.com/skills/agensi-benchmarking-ai-agents-beyond-models/install.json","trustJson":"https://agentstaffer.com/skills/agensi-benchmarking-ai-agents-beyond-models/trust.json","trialJson":"https://agentstaffer.com/skills/agensi-benchmarking-ai-agents-beyond-models/trial.json"},"queueRank":13,"recommendation":{"label":"Needs review","score":37},"trialPriority":57,"trial":{"title":"Run a small code review trial","input":"Give the agent one low-risk issue or one small pull request and ask for defects, tests, and a patch plan.","expectedOutput":"A concise review with reproducible findings, changed-file scope, and test suggestions.","safetyNotes":["Run in a sandbox or throwaway workspace.","Use non-production data and temporary credentials.","Capture output, logs, install friction, and any permission requests."],"decisionCriteria":["Finds concrete issues","Avoids broad rewrites","Explains tradeoffs","Can be evaluated without secrets"],"runUrl":"https://agentstaffer.com/skills/agensi-benchmarking-ai-agents-beyond-models#trial","runPath":"/skills/agensi-benchmarking-ai-agents-beyond-models#trial","createEndpoint":"https://agentstaffer.com/api/actions/run-trial","createBody":{"listingId":"agensi:skills-benchmarking-ai-agents-beyond-models","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}},"launchPacket":{"schemaVersion":"agentstaffer.trial-launch-packet-lite.v1","requiredPermissions":{"schemaVersion":"agentstaffer.trial-requirements.v1","principle":"Use the smallest sandbox permission set needed for the first reversible trial.","requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required","text":"Sandbox workspace: Run the first trial in a throwaway directory or isolated environment."},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review","text":"License review: License metadata is missing from the current public profile."}],"humanReadable":["Sandbox workspace: Run the first trial in a throwaway directory or isolated environment.","License review: License metadata is missing from the current public profile."]},"preflightGates":[{"id":"profile","label":"Profile record","href":"https://agentstaffer.com/skills/agensi-benchmarking-ai-agents-beyond-models/profile.json","passed":true},{"id":"install","label":"Setup guide","href":"https://agentstaffer.com/skills/agensi-benchmarking-ai-agents-beyond-models/install.json","passed":true},{"id":"sandbox","label":"Use sandbox data","passed":true}],"promotionGates":[{"id":"install","label":"Install path reviewed","passed":true},{"id":"trust","label":"Trust evidence reviewed","passed":true},{"id":"trial","label":"Sandbox trial recorded","passed":false}],"stopConditions":["Install asks for production credentials or broad write access.","Source or package cannot be verified from profile evidence."],"createRequest":{"endpoint":"https://agentstaffer.com/api/actions/run-trial","method":"POST","body":{"listingId":"agensi:skills-benchmarking-ai-agents-beyond-models","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}}},"safety":{"trustTier":"Verified source","installReadiness":"Source guide","topRisk":{"level":"medium","label":"No community evaluation","detail":"AgentStaffer users have not submitted a review yet."},"requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required"},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review"}],"permissionSummary":{"schemaVersion":"agentstaffer.trial-requirements.v1","principle":"Use the smallest sandbox permission set needed for the first reversible trial.","requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required","text":"Sandbox workspace: Run the first trial in a throwaway directory or isolated environment."},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review","text":"License review: License metadata is missing from the current public profile."}],"humanReadable":["Sandbox workspace: Run the first trial in a throwaway directory or isolated environment.","License review: License metadata is missing from the current public profile."]}}},{"id":"remoteopenclaw:skills-sickn33-antigravity-awesome-skills-agent-evaluation","slug":"remoteopenclaw-agent-evaluation","name":"agent-evaluation","type":"skill","source":"remoteopenclaw","status":"pending","summary":"Testing and benchmarking LLM agents including behavioral testing,","profileUrl":"https://agentstaffer.com/skills/remoteopenclaw-agent-evaluation","fitScore":37,"roleFamily":"Coding & Development","reviewSignal":"Pending","urls":{"profileJson":"https://agentstaffer.com/skills/remoteopenclaw-agent-evaluation/profile.json","installJson":"https://agentstaffer.com/skills/remoteopenclaw-agent-evaluation/install.json","trustJson":"https://agentstaffer.com/skills/remoteopenclaw-agent-evaluation/trust.json","trialJson":"https://agentstaffer.com/skills/remoteopenclaw-agent-evaluation/trial.json"},"queueRank":14,"recommendation":{"label":"Needs review","score":37},"trialPriority":57,"trial":{"title":"Run a small code review trial","input":"Give the agent one low-risk issue or one small pull request and ask for defects, tests, and a patch plan.","expectedOutput":"A concise review with reproducible findings, changed-file scope, and test suggestions.","safetyNotes":["Run in a sandbox or throwaway workspace.","Use non-production data and temporary credentials.","Capture output, logs, install friction, and any permission requests."],"decisionCriteria":["Finds concrete issues","Avoids broad rewrites","Explains tradeoffs","Can be evaluated without secrets"],"runUrl":"https://agentstaffer.com/skills/remoteopenclaw-agent-evaluation#trial","runPath":"/skills/remoteopenclaw-agent-evaluation#trial","createEndpoint":"https://agentstaffer.com/api/actions/run-trial","createBody":{"listingId":"remoteopenclaw:skills-sickn33-antigravity-awesome-skills-agent-evaluation","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}},"launchPacket":{"schemaVersion":"agentstaffer.trial-launch-packet-lite.v1","requiredPermissions":{"schemaVersion":"agentstaffer.trial-requirements.v1","principle":"Use the smallest sandbox permission set needed for the first reversible trial.","requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required","text":"Sandbox workspace: Run the first trial in a throwaway directory or isolated environment."},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review","text":"License review: License metadata is missing from the current public profile."}],"humanReadable":["Sandbox workspace: Run the first trial in a throwaway directory or isolated environment.","License review: License metadata is missing from the current public profile."]},"preflightGates":[{"id":"profile","label":"Profile record","href":"https://agentstaffer.com/skills/remoteopenclaw-agent-evaluation/profile.json","passed":true},{"id":"install","label":"Setup guide","href":"https://agentstaffer.com/skills/remoteopenclaw-agent-evaluation/install.json","passed":true},{"id":"sandbox","label":"Use sandbox data","passed":true}],"promotionGates":[{"id":"install","label":"Install path reviewed","passed":true},{"id":"trust","label":"Trust evidence reviewed","passed":false},{"id":"trial","label":"Sandbox trial recorded","passed":false}],"stopConditions":["Install asks for production credentials or broad write access.","Source or package cannot be verified from profile evidence."],"createRequest":{"endpoint":"https://agentstaffer.com/api/actions/run-trial","method":"POST","body":{"listingId":"remoteopenclaw:skills-sickn33-antigravity-awesome-skills-agent-evaluation","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}}},"safety":{"trustTier":"Needs review","installReadiness":"Source guide","topRisk":{"level":"medium","label":"No community evaluation","detail":"AgentStaffer users have not submitted a review yet."},"requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required"},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review"}],"permissionSummary":{"schemaVersion":"agentstaffer.trial-requirements.v1","principle":"Use the smallest sandbox permission set needed for the first reversible trial.","requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required","text":"Sandbox workspace: Run the first trial in a throwaway directory or isolated environment."},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review","text":"License review: License metadata is missing from the current public profile."}],"humanReadable":["Sandbox workspace: Run the first trial in a throwaway directory or isolated environment.","License review: License metadata is missing from the current public profile."]}}},{"id":"remoteopenclaw:skills-davila7-claude-code-templates-agent-evaluation","slug":"remoteopenclaw-agent-evaluation","name":"agent-evaluation","type":"skill","source":"remoteopenclaw","status":"pending","summary":"Testing and benchmarking LLM agents including behavioral testing, capability assessment, reliability metrics, and production monitoring—where even top agents achieve less than 50% on real-world benchmarks Use when: agent testing, agent evaluation, benchmark agents, agent reliability, test agent.","profileUrl":"https://agentstaffer.com/skills/remoteopenclaw-agent-evaluation","fitScore":37,"roleFamily":"Coding & Development","reviewSignal":"Pending","urls":{"profileJson":"https://agentstaffer.com/skills/remoteopenclaw-agent-evaluation/profile.json","installJson":"https://agentstaffer.com/skills/remoteopenclaw-agent-evaluation/install.json","trustJson":"https://agentstaffer.com/skills/remoteopenclaw-agent-evaluation/trust.json","trialJson":"https://agentstaffer.com/skills/remoteopenclaw-agent-evaluation/trial.json"},"queueRank":15,"recommendation":{"label":"Needs review","score":37},"trialPriority":57,"trial":{"title":"Run a small code review trial","input":"Give the agent one low-risk issue or one small pull request and ask for defects, tests, and a patch plan.","expectedOutput":"A concise review with reproducible findings, changed-file scope, and test suggestions.","safetyNotes":["Run in a sandbox or throwaway workspace.","Use non-production data and temporary credentials.","Capture output, logs, install friction, and any permission requests."],"decisionCriteria":["Finds concrete issues","Avoids broad rewrites","Explains tradeoffs","Can be evaluated without secrets"],"runUrl":"https://agentstaffer.com/skills/remoteopenclaw-agent-evaluation#trial","runPath":"/skills/remoteopenclaw-agent-evaluation#trial","createEndpoint":"https://agentstaffer.com/api/actions/run-trial","createBody":{"listingId":"remoteopenclaw:skills-davila7-claude-code-templates-agent-evaluation","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}},"launchPacket":{"schemaVersion":"agentstaffer.trial-launch-packet-lite.v1","requiredPermissions":{"schemaVersion":"agentstaffer.trial-requirements.v1","principle":"Use the smallest sandbox permission set needed for the first reversible trial.","requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required","text":"Sandbox workspace: Run the first trial in a throwaway directory or isolated environment."},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review","text":"License review: License metadata is missing from the current public profile."}],"humanReadable":["Sandbox workspace: Run the first trial in a throwaway directory or isolated environment.","License review: License metadata is missing from the current public profile."]},"preflightGates":[{"id":"profile","label":"Profile record","href":"https://agentstaffer.com/skills/remoteopenclaw-agent-evaluation/profile.json","passed":true},{"id":"install","label":"Setup guide","href":"https://agentstaffer.com/skills/remoteopenclaw-agent-evaluation/install.json","passed":true},{"id":"sandbox","label":"Use sandbox data","passed":true}],"promotionGates":[{"id":"install","label":"Install path reviewed","passed":true},{"id":"trust","label":"Trust evidence reviewed","passed":false},{"id":"trial","label":"Sandbox trial recorded","passed":false}],"stopConditions":["Install asks for production credentials or broad write access.","Source or package cannot be verified from profile evidence."],"createRequest":{"endpoint":"https://agentstaffer.com/api/actions/run-trial","method":"POST","body":{"listingId":"remoteopenclaw:skills-davila7-claude-code-templates-agent-evaluation","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}}},"safety":{"trustTier":"Needs review","installReadiness":"Source guide","topRisk":{"level":"medium","label":"No community evaluation","detail":"AgentStaffer users have not submitted a review yet."},"requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required"},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review"}],"permissionSummary":{"schemaVersion":"agentstaffer.trial-requirements.v1","principle":"Use the smallest sandbox permission set needed for the first reversible trial.","requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required","text":"Sandbox workspace: Run the first trial in a throwaway directory or isolated environment."},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review","text":"License review: License metadata is missing from the current public profile."}],"humanReadable":["Sandbox workspace: Run the first trial in a throwaway directory or isolated environment.","License review: License metadata is missing from the current public profile."]}}},{"id":"mcpregistry:servers-io-github-verifyax-2fmcp-server","slug":"mcpregistry-verifyax","name":"VerifyAX","type":"tool","source":"mcpregistry","status":"pending","summary":"Evaluate, benchmark, and simulate AI agents on the VerifyAX agent-evaluation platform.","profileUrl":"https://agentstaffer.com/mcp/mcpregistry-verifyax","fitScore":38,"roleFamily":"Workflow Automation","reviewSignal":"Pending","urls":{"profileJson":"https://agentstaffer.com/mcp/mcpregistry-verifyax/profile.json","installJson":"https://agentstaffer.com/mcp/mcpregistry-verifyax/install.json","trustJson":"https://agentstaffer.com/mcp/mcpregistry-verifyax/trust.json","trialJson":"https://agentstaffer.com/mcp/mcpregistry-verifyax/trial.json"},"queueRank":16,"recommendation":{"label":"Needs review","score":38},"trialPriority":58,"trial":{"title":"Run a dry-run workflow trial","input":"Connect the workflow to mock inputs and ask it to produce a trace without touching production systems.","expectedOutput":"A step-by-step run log with inputs, outputs, failures, and rollback notes.","safetyNotes":["Run in a sandbox or throwaway workspace.","Use non-production data and temporary credentials.","Capture output, logs, install friction, and any permission requests."],"decisionCriteria":["Shows observable steps","Handles failure paths","Documents required credentials","Can run without side effects"],"runUrl":"https://agentstaffer.com/mcp/mcpregistry-verifyax#trial","runPath":"/mcp/mcpregistry-verifyax#trial","createEndpoint":"https://agentstaffer.com/api/actions/run-trial","createBody":{"listingId":"mcpregistry:servers-io-github-verifyax-2fmcp-server","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}},"launchPacket":{"schemaVersion":"agentstaffer.trial-launch-packet-lite.v1","requiredPermissions":{"schemaVersion":"agentstaffer.trial-requirements.v1","principle":"Use the smallest sandbox permission set needed for the first reversible trial.","requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required","text":"Sandbox workspace: Run the first trial in a throwaway directory or isolated environment."},{"id":"mcp-client","label":"MCP-capable client","detail":"Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","severity":"required","text":"MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs."},{"id":"temporary-credentials","label":"Temporary credentials","detail":"Use throwaway API keys or scoped test credentials. Do not start with production secrets.","severity":"conditional","text":"Temporary credentials: Use throwaway API keys or scoped test credentials. Do not start with production secrets."},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review","text":"License review: License metadata is missing from the current public profile."}],"humanReadable":["Sandbox workspace: Run the first trial in a throwaway directory or isolated environment.","MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","Temporary credentials: Use throwaway API keys or scoped test credentials. Do not start with production secrets.","License review: License metadata is missing from the current public profile."]},"preflightGates":[{"id":"profile","label":"Profile record","href":"https://agentstaffer.com/mcp/mcpregistry-verifyax/profile.json","passed":true},{"id":"install","label":"Setup guide","href":"https://agentstaffer.com/mcp/mcpregistry-verifyax/install.json","passed":true},{"id":"sandbox","label":"Use sandbox data","passed":true}],"promotionGates":[{"id":"install","label":"Install path reviewed","passed":true},{"id":"trust","label":"Trust evidence reviewed","passed":false},{"id":"trial","label":"Sandbox trial recorded","passed":false}],"stopConditions":["Install asks for production credentials or broad write access.","Source or package cannot be verified from profile evidence."],"createRequest":{"endpoint":"https://agentstaffer.com/api/actions/run-trial","method":"POST","body":{"listingId":"mcpregistry:servers-io-github-verifyax-2fmcp-server","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}}},"safety":{"trustTier":"Needs review","installReadiness":"MCP ready","topRisk":{"level":"medium","label":"No community evaluation","detail":"AgentStaffer users have not submitted a review yet."},"requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required"},{"id":"mcp-client","label":"MCP-capable client","detail":"Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","severity":"required"},{"id":"temporary-credentials","label":"Temporary credentials","detail":"Use throwaway API keys or scoped test credentials. Do not start with production secrets.","severity":"conditional"},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review"}],"permissionSummary":{"schemaVersion":"agentstaffer.trial-requirements.v1","principle":"Use the smallest sandbox permission set needed for the first reversible trial.","requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required","text":"Sandbox workspace: Run the first trial in a throwaway directory or isolated environment."},{"id":"mcp-client","label":"MCP-capable client","detail":"Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","severity":"required","text":"MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs."},{"id":"temporary-credentials","label":"Temporary credentials","detail":"Use throwaway API keys or scoped test credentials. Do not start with production secrets.","severity":"conditional","text":"Temporary credentials: Use throwaway API keys or scoped test credentials. Do not start with production secrets."},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review","text":"License review: License metadata is missing from the current public profile."}],"humanReadable":["Sandbox workspace: Run the first trial in a throwaway directory or isolated environment.","MCP-capable client: Use Claude Desktop, Cursor, VS Code, or another client that can load MCP server configs.","Temporary credentials: Use throwaway API keys or scoped test credentials. Do not start with production secrets.","License review: License metadata is missing from the current public profile."]}}},{"id":"awesomeagents:gauravsbin-cognitive-security-ai-powered-threat-agent-evaluation-for-impact-on-assets-","slug":"awesomeagents-cognitive-security-ai-powered-threat-agent-evaluation-for-impact-on-assets","name":"Cognitive-Security-Ai-Powered-Threat-Agent-Evaluation-For-Impact-On-Assets.","type":"tool","source":"awesomeagents","status":"pending","summary":"No description available","profileUrl":"https://agentstaffer.com/agents/awesomeagents-cognitive-security-ai-powered-threat-agent-evaluation-for-impact-on-assets","fitScore":27,"roleFamily":"Workflow Automation","reviewSignal":"Pending","urls":{"profileJson":"https://agentstaffer.com/agents/awesomeagents-cognitive-security-ai-powered-threat-agent-evaluation-for-impact-on-assets/profile.json","installJson":"https://agentstaffer.com/agents/awesomeagents-cognitive-security-ai-powered-threat-agent-evaluation-for-impact-on-assets/install.json","trustJson":"https://agentstaffer.com/agents/awesomeagents-cognitive-security-ai-powered-threat-agent-evaluation-for-impact-on-assets/trust.json","trialJson":"https://agentstaffer.com/agents/awesomeagents-cognitive-security-ai-powered-threat-agent-evaluation-for-impact-on-assets/trial.json"},"queueRank":17,"recommendation":{"label":"Needs review","score":27},"trialPriority":47,"trial":{"title":"Run a dry-run workflow trial","input":"Connect the workflow to mock inputs and ask it to produce a trace without touching production systems.","expectedOutput":"A step-by-step run log with inputs, outputs, failures, and rollback notes.","safetyNotes":["Run in a sandbox or throwaway workspace.","Use non-production data and temporary credentials.","Capture output, logs, install friction, and any permission requests."],"decisionCriteria":["Shows observable steps","Handles failure paths","Documents required credentials","Can run without side effects"],"runUrl":"https://agentstaffer.com/agents/awesomeagents-cognitive-security-ai-powered-threat-agent-evaluation-for-impact-on-assets#trial","runPath":"/agents/awesomeagents-cognitive-security-ai-powered-threat-agent-evaluation-for-impact-on-assets#trial","createEndpoint":"https://agentstaffer.com/api/actions/run-trial","createBody":{"listingId":"awesomeagents:gauravsbin-cognitive-security-ai-powered-threat-agent-evaluation-for-impact-on-assets-","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}},"launchPacket":{"schemaVersion":"agentstaffer.trial-launch-packet-lite.v1","requiredPermissions":{"schemaVersion":"agentstaffer.trial-requirements.v1","principle":"Use the smallest sandbox permission set needed for the first reversible trial.","requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required","text":"Sandbox workspace: Run the first trial in a throwaway directory or isolated environment."},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review","text":"License review: License metadata is missing from the current public profile."}],"humanReadable":["Sandbox workspace: Run the first trial in a throwaway directory or isolated environment.","License review: License metadata is missing from the current public profile."]},"preflightGates":[{"id":"profile","label":"Profile record","href":"https://agentstaffer.com/agents/awesomeagents-cognitive-security-ai-powered-threat-agent-evaluation-for-impact-on-assets/profile.json","passed":true},{"id":"install","label":"Setup guide","href":"https://agentstaffer.com/agents/awesomeagents-cognitive-security-ai-powered-threat-agent-evaluation-for-impact-on-assets/install.json","passed":true},{"id":"sandbox","label":"Use sandbox data","passed":true}],"promotionGates":[{"id":"install","label":"Install path reviewed","passed":true},{"id":"trust","label":"Trust evidence reviewed","passed":false},{"id":"trial","label":"Sandbox trial recorded","passed":false}],"stopConditions":["Install asks for production credentials or broad write access.","Source or package cannot be verified from profile evidence."],"createRequest":{"endpoint":"https://agentstaffer.com/api/actions/run-trial","method":"POST","body":{"listingId":"awesomeagents:gauravsbin-cognitive-security-ai-powered-threat-agent-evaluation-for-impact-on-assets-","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}}},"safety":{"trustTier":"Needs review","installReadiness":"Source guide","topRisk":{"level":"medium","label":"No community evaluation","detail":"AgentStaffer users have not submitted a review yet."},"requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required"},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review"}],"permissionSummary":{"schemaVersion":"agentstaffer.trial-requirements.v1","principle":"Use the smallest sandbox permission set needed for the first reversible trial.","requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required","text":"Sandbox workspace: Run the first trial in a throwaway directory or isolated environment."},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review","text":"License review: License metadata is missing from the current public profile."}],"humanReadable":["Sandbox workspace: Run the first trial in a throwaway directory or isolated environment.","License review: License metadata is missing from the current public profile."]}}},{"id":"awesomeagents:awslabs-agent-evaluation","slug":"awesomeagents-agent-evaluation","name":"Agent-Evaluation","type":"framework","source":"awesomeagents","status":"pending","summary":"A generative AI-powered framework for testing virtual agents.","profileUrl":"https://agentstaffer.com/agents/awesomeagents-agent-evaluation","fitScore":27,"roleFamily":"Workflow Automation","reviewSignal":"Pending","urls":{"profileJson":"https://agentstaffer.com/agents/awesomeagents-agent-evaluation/profile.json","installJson":"https://agentstaffer.com/agents/awesomeagents-agent-evaluation/install.json","trustJson":"https://agentstaffer.com/agents/awesomeagents-agent-evaluation/trust.json","trialJson":"https://agentstaffer.com/agents/awesomeagents-agent-evaluation/trial.json"},"queueRank":18,"recommendation":{"label":"Needs review","score":27},"trialPriority":47,"trial":{"title":"Run a dry-run workflow trial","input":"Connect the workflow to mock inputs and ask it to produce a trace without touching production systems.","expectedOutput":"A step-by-step run log with inputs, outputs, failures, and rollback notes.","safetyNotes":["Run in a sandbox or throwaway workspace.","Use non-production data and temporary credentials.","Capture output, logs, install friction, and any permission requests."],"decisionCriteria":["Shows observable steps","Handles failure paths","Documents required credentials","Can run without side effects"],"runUrl":"https://agentstaffer.com/agents/awesomeagents-agent-evaluation#trial","runPath":"/agents/awesomeagents-agent-evaluation#trial","createEndpoint":"https://agentstaffer.com/api/actions/run-trial","createBody":{"listingId":"awesomeagents:awslabs-agent-evaluation","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}},"launchPacket":{"schemaVersion":"agentstaffer.trial-launch-packet-lite.v1","requiredPermissions":{"schemaVersion":"agentstaffer.trial-requirements.v1","principle":"Use the smallest sandbox permission set needed for the first reversible trial.","requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required","text":"Sandbox workspace: Run the first trial in a throwaway directory or isolated environment."},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review","text":"License review: License metadata is missing from the current public profile."}],"humanReadable":["Sandbox workspace: Run the first trial in a throwaway directory or isolated environment.","License review: License metadata is missing from the current public profile."]},"preflightGates":[{"id":"profile","label":"Profile record","href":"https://agentstaffer.com/agents/awesomeagents-agent-evaluation/profile.json","passed":true},{"id":"install","label":"Setup guide","href":"https://agentstaffer.com/agents/awesomeagents-agent-evaluation/install.json","passed":true},{"id":"sandbox","label":"Use sandbox data","passed":true}],"promotionGates":[{"id":"install","label":"Install path reviewed","passed":true},{"id":"trust","label":"Trust evidence reviewed","passed":false},{"id":"trial","label":"Sandbox trial recorded","passed":false}],"stopConditions":["Install asks for production credentials or broad write access.","Source or package cannot be verified from profile evidence."],"createRequest":{"endpoint":"https://agentstaffer.com/api/actions/run-trial","method":"POST","body":{"listingId":"awesomeagents:awslabs-agent-evaluation","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}}},"safety":{"trustTier":"Needs review","installReadiness":"Source guide","topRisk":{"level":"medium","label":"No community evaluation","detail":"AgentStaffer users have not submitted a review yet."},"requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required"},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review"}],"permissionSummary":{"schemaVersion":"agentstaffer.trial-requirements.v1","principle":"Use the smallest sandbox permission set needed for the first reversible trial.","requirements":[{"id":"sandbox","label":"Sandbox workspace","detail":"Run the first trial in a throwaway directory or isolated environment.","severity":"required","text":"Sandbox workspace: Run the first trial in a throwaway directory or isolated environment."},{"id":"license-review","label":"License review","detail":"License metadata is missing from the current public profile.","severity":"review","text":"License review: License metadata is missing from the current public profile."}],"humanReadable":["Sandbox workspace: Run the first trial in a throwaway directory or isolated environment.","License review: License metadata is missing from the current public profile."]}}}],"recentRuns":[],"pagination":{"page":1,"pageSize":24,"totalItems":18,"totalPages":1,"nextPageUrl":null,"prevPageUrl":null},"filters":{"q":"agent-evaluation","type":null,"role":null,"protocol":null,"trust":null,"install":null,"source":null,"status":null,"sort":"trial_readiness"},"availableFilters":{"roles":[{"id":"coding-development","label":"Coding & Development"},{"id":"workflow-automation","label":"Workflow Automation"},{"id":"multi-agent-orchestration","label":"Multi-agent Orchestration"},{"id":"mcp-tool-integration","label":"MCP & Tool Integration"},{"id":"research-knowledge","label":"Research & Knowledge"},{"id":"computer-use-desktop","label":"Computer Use & Desktop"},{"id":"business-operations","label":"Business Operations"},{"id":"data-integrations","label":"Data & Integrations"},{"id":"evaluation-safety","label":"Evaluation & Safety"},{"id":"model-inference","label":"Models & Inference"},{"id":"general-agent","label":"General Agent"}],"protocols":[{"value":"mcp","label":"MCP (10)"},{"value":"cli","label":"CLI (8)"},{"value":"http","label":"HTTP (5)"},{"value":"sdk","label":"SDK (3)"}],"trust":[{"value":"needs-review","label":"Needs review (17)"},{"value":"verified-source","label":"Verified source (1)"}],"install":[{"value":"mcp-ready","label":"MCP ready (10)"},{"value":"source-guide","label":"Source guide (7)"},{"value":"cli-ready","label":"CLI ready (1)"}],"sources":[{"value":"playbooks","label":"playbooks (5)"},{"value":"remoteopenclaw","label":"remoteopenclaw (3)"},{"value":"awesomeagents","label":"awesomeagents (2)"},{"value":"mcpregistry","label":"mcpregistry (2)"},{"value":"npm","label":"npm (2)"},{"value":"agensi","label":"agensi (1)"},{"value":"claudemarketplaces","label":"claudemarketplaces (1)"},{"value":"explainx","label":"explainx (1)"},{"value":"mcpservers","label":"mcpservers (1)"}],"types":[{"value":"skill","label":"skill (11)"},{"value":"tool","label":"tool (4)"},{"value":"workflow","label":"workflow (2)"},{"value":"framework","label":"framework (1)"}],"sortModes":["relevance","trust","updated","popularity","install_readiness","trial_readiness"]},"launchProtocol":{"schemaVersion":"agentstaffer.trial-launch-packet.v1","location":"trialQueue[].launchPacket","permissionsLocation":"trialQueue[].safety.permissionSummary","rule":"Use launchPacket.preflightGates and launchPacket.stopConditions before calling launchPacket.createRequest.","promotionRule":"Do not promote or hire until launchPacket.promotionGates are clear or explicitly accepted by a human reviewer."},"visibleRetrievalSpine":[{"step":"01","label":"llms.txt","href":"/llms.txt","note":"Site policy and safety context"},{"step":"02","label":"trials.json","href":"/trials.json?q=agent-evaluation&sort=trial_readiness","note":"Action queue and page contract"},{"step":"03","label":"profile HTML","href":"/mcp/npm-awaithumans","note":"Human decision summary"},{"step":"04","label":"profile package","href":"/mcp/npm-awaithumans/profile.json","note":"Fetch all 9 required finalist resources","requiredResources":["profile.json","score.json","evidence.json","trust.json","trial.json","install.json","history.json","similar.json","dossier.md"]}],"profileResourceContract":{"schemaVersion":"agentstaffer.profile-resource-contract.v1","requiredResources":["profile.json","score.json","evidence.json","trust.json","trial.json","install.json","history.json","similar.json","dossier.md"],"recommendedFetchOrder":["profile.json","score.json","evidence.json","trust.json","trial.json","install.json","history.json","similar.json","dossier.md"]},"retrievalHints":["Use trialQueue[].trial before recommending hire or production installation.","Use trialQueue[].safety.permissionSummary.humanReadable when explaining first-run permissions to a human.","Use trialQueue[].launchPacket as the copyable trial launch contract for humans and AI agents.","POST to /api/actions/run-trial with trialQueue[].trial.createBody to create a public trial result.","Fetch trialQueue[].urls.profileJson (profile.json) and trialQueue[].urls.installJson (install.json) for the full dossier and install payloads.","Use recentRuns[] as auditable evidence of completed or blocked sandbox evaluations."],"links":{"trials":"https://agentstaffer.com/trials?q=agent-evaluation&sort=trial_readiness","trialsJson":"https://agentstaffer.com/trials.json?q=agent-evaluation&sort=trial_readiness","compare":"https://agentstaffer.com/compare","promotions":"https://agentstaffer.com/promotions","jobMatcher":"https://agentstaffer.com/job-matcher","openApi":"https://agentstaffer.com/openapi.json","llmsTxt":"https://agentstaffer.com/llms.txt","geoJson":"https://agentstaffer.com/geo.json","agentVisibilityJson":"https://agentstaffer.com/agent-visibility.json","aiCrawlMonitorJson":"https://agentstaffer.com/ai-crawl-monitor.json"}}