{"schemaVersion":"agentstaffer.trial-queue.v0","generatedAt":"2026-10-17T15:44:20.516Z","canonicalUrl":"https://agentstaffer.com/trials.json?q=Humane&sort=trial_readiness","htmlUrl":"https://agentstaffer.com/trials?q=Humane&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":16,"trialQueue":[{"id":"explainx:agents-humane","slug":"explainx-humane","name":"Humane","type":"agent","source":"explainx","status":"pending","summary":"See the world, not your screen.","profileUrl":"https://agentstaffer.com/mcp/explainx-humane","fitScore":61,"roleFamily":"Coding & Development","reviewSignal":"Pending","urls":{"profileJson":"https://agentstaffer.com/mcp/explainx-humane/profile.json","installJson":"https://agentstaffer.com/mcp/explainx-humane/install.json","trustJson":"https://agentstaffer.com/mcp/explainx-humane/trust.json","trialJson":"https://agentstaffer.com/mcp/explainx-humane/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/explainx-humane#trial","runPath":"/mcp/explainx-humane#trial","createEndpoint":"https://agentstaffer.com/api/actions/run-trial","createBody":{"listingId":"explainx:agents-humane","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/explainx-humane/profile.json","passed":true},{"id":"install","label":"Setup guide","href":"https://agentstaffer.com/mcp/explainx-humane/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:agents-humane","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-coowoolf-insighthunt-skills-agent-mindset-termination","slug":"playbooks-agent-mindset-termination-skill-by-coowoolf-insighthunt-skills","name":"agent-mindset-termination skill by coowoolf/insighthunt-skills","type":"skill","source":"playbooks","status":"pending","summary":"This skill helps managers handle tough terminations humanely while actively supporting the departing employee in finding their next role.","profileUrl":"https://agentstaffer.com/skills/playbooks-agent-mindset-termination-skill-by-coowoolf-insighthunt-skills","fitScore":55,"roleFamily":"Workflow Automation","reviewSignal":"Pending","urls":{"profileJson":"https://agentstaffer.com/skills/playbooks-agent-mindset-termination-skill-by-coowoolf-insighthunt-skills/profile.json","installJson":"https://agentstaffer.com/skills/playbooks-agent-mindset-termination-skill-by-coowoolf-insighthunt-skills/install.json","trustJson":"https://agentstaffer.com/skills/playbooks-agent-mindset-termination-skill-by-coowoolf-insighthunt-skills/trust.json","trialJson":"https://agentstaffer.com/skills/playbooks-agent-mindset-termination-skill-by-coowoolf-insighthunt-skills/trial.json"},"queueRank":2,"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-mindset-termination-skill-by-coowoolf-insighthunt-skills#trial","runPath":"/skills/playbooks-agent-mindset-termination-skill-by-coowoolf-insighthunt-skills#trial","createEndpoint":"https://agentstaffer.com/api/actions/run-trial","createBody":{"listingId":"playbooks:skills-coowoolf-insighthunt-skills-agent-mindset-termination","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-mindset-termination-skill-by-coowoolf-insighthunt-skills/profile.json","passed":true},{"id":"install","label":"Setup guide","href":"https://agentstaffer.com/skills/playbooks-agent-mindset-termination-skill-by-coowoolf-insighthunt-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-coowoolf-insighthunt-skills-agent-mindset-termination","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:agent-skills-firecrawl-nemo-evaluator-sdk","slug":"mcpservers-nemo-evaluator-sdk","name":"nemo-evaluator-sdk","type":"skill","source":"mcpservers","status":"pending","summary":"Evaluates LLMs across 100+ benchmarks from 18+ harnesses (MMLU, HumanEval, GSM8K, safety, VLM) with multi-backend execution. Use when needing scalable…","profileUrl":"https://agentstaffer.com/skills/mcpservers-nemo-evaluator-sdk","fitScore":66,"roleFamily":"Coding & Development","reviewSignal":"Pending","urls":{"profileJson":"https://agentstaffer.com/skills/mcpservers-nemo-evaluator-sdk/profile.json","installJson":"https://agentstaffer.com/skills/mcpservers-nemo-evaluator-sdk/install.json","trustJson":"https://agentstaffer.com/skills/mcpservers-nemo-evaluator-sdk/trust.json","trialJson":"https://agentstaffer.com/skills/mcpservers-nemo-evaluator-sdk/trial.json"},"queueRank":3,"recommendation":{"label":"Needs review","score":66},"trialPriority":86,"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/mcpservers-nemo-evaluator-sdk#trial","runPath":"/skills/mcpservers-nemo-evaluator-sdk#trial","createEndpoint":"https://agentstaffer.com/api/actions/run-trial","createBody":{"listingId":"mcpservers:agent-skills-firecrawl-nemo-evaluator-sdk","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/mcpservers-nemo-evaluator-sdk/profile.json","passed":true},{"id":"install","label":"Setup guide","href":"https://agentstaffer.com/skills/mcpservers-nemo-evaluator-sdk/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:agent-skills-firecrawl-nemo-evaluator-sdk","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-linkedin-human-warmup","slug":"playbooks-linkedin-human-warmup-skill-by-openclaw-skills","name":"linkedin-human-warmup skill by openclaw/skills","type":"skill","source":"playbooks","status":"pending","summary":"This skill guides humane LinkedIn warmup for new or cold accounts by planning memory-informed scripts and risk-aware actions.","profileUrl":"https://agentstaffer.com/skills/playbooks-linkedin-human-warmup-skill-by-openclaw-skills","fitScore":55,"roleFamily":"Coding & Development","reviewSignal":"Pending","urls":{"profileJson":"https://agentstaffer.com/skills/playbooks-linkedin-human-warmup-skill-by-openclaw-skills/profile.json","installJson":"https://agentstaffer.com/skills/playbooks-linkedin-human-warmup-skill-by-openclaw-skills/install.json","trustJson":"https://agentstaffer.com/skills/playbooks-linkedin-human-warmup-skill-by-openclaw-skills/trust.json","trialJson":"https://agentstaffer.com/skills/playbooks-linkedin-human-warmup-skill-by-openclaw-skills/trial.json"},"queueRank":4,"recommendation":{"label":"Needs review","score":55},"trialPriority":75,"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-linkedin-human-warmup-skill-by-openclaw-skills#trial","runPath":"/skills/playbooks-linkedin-human-warmup-skill-by-openclaw-skills#trial","createEndpoint":"https://agentstaffer.com/api/actions/run-trial","createBody":{"listingId":"playbooks:skills-openclaw-skills-linkedin-human-warmup","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-linkedin-human-warmup-skill-by-openclaw-skills/profile.json","passed":true},{"id":"install","label":"Setup guide","href":"https://agentstaffer.com/skills/playbooks-linkedin-human-warmup-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-linkedin-human-warmup","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:agent-skills-firecrawl-evaluating-code-models","slug":"mcpservers-evaluating-code-models","name":"evaluating-code-models","type":"skill","source":"mcpservers","status":"pending","summary":"Evaluates code generation models across HumanEval, MBPP, MultiPL-E, and 15+ benchmarks with pass@k metrics. Use when benchmarking code models, comparing coding…","profileUrl":"https://agentstaffer.com/skills/mcpservers-evaluating-code-models","fitScore":61,"roleFamily":"Coding & Development","reviewSignal":"Pending","urls":{"profileJson":"https://agentstaffer.com/skills/mcpservers-evaluating-code-models/profile.json","installJson":"https://agentstaffer.com/skills/mcpservers-evaluating-code-models/install.json","trustJson":"https://agentstaffer.com/skills/mcpservers-evaluating-code-models/trust.json","trialJson":"https://agentstaffer.com/skills/mcpservers-evaluating-code-models/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/mcpservers-evaluating-code-models#trial","runPath":"/skills/mcpservers-evaluating-code-models#trial","createEndpoint":"https://agentstaffer.com/api/actions/run-trial","createBody":{"listingId":"mcpservers:agent-skills-firecrawl-evaluating-code-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":"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/mcpservers-evaluating-code-models/profile.json","passed":true},{"id":"install","label":"Setup guide","href":"https://agentstaffer.com/skills/mcpservers-evaluating-code-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":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:agent-skills-firecrawl-evaluating-code-models","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:agent-skills-firecrawl-evaluating-llms-harness","slug":"mcpservers-evaluating-llms-harness","name":"evaluating-llms-harness","type":"skill","source":"mcpservers","status":"pending","summary":"Evaluates LLMs across 60+ academic benchmarks (MMLU, HumanEval, GSM8K, TruthfulQA, HellaSwag). Use when benchmarking model quality, comparing models, reporting…","profileUrl":"https://agentstaffer.com/skills/mcpservers-evaluating-llms-harness","fitScore":61,"roleFamily":"Coding & Development","reviewSignal":"Pending","urls":{"profileJson":"https://agentstaffer.com/skills/mcpservers-evaluating-llms-harness/profile.json","installJson":"https://agentstaffer.com/skills/mcpservers-evaluating-llms-harness/install.json","trustJson":"https://agentstaffer.com/skills/mcpservers-evaluating-llms-harness/trust.json","trialJson":"https://agentstaffer.com/skills/mcpservers-evaluating-llms-harness/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/mcpservers-evaluating-llms-harness#trial","runPath":"/skills/mcpservers-evaluating-llms-harness#trial","createEndpoint":"https://agentstaffer.com/api/actions/run-trial","createBody":{"listingId":"mcpservers:agent-skills-firecrawl-evaluating-llms-harness","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/mcpservers-evaluating-llms-harness/profile.json","passed":true},{"id":"install","label":"Setup guide","href":"https://agentstaffer.com/skills/mcpservers-evaluating-llms-harness/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:agent-skills-firecrawl-evaluating-llms-harness","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":"aiagentsdirectory:agent-humane-ai-pin","slug":"aiagentsdirectory-humane-ai-pin","name":"Humane AI Pin","type":"agent","source":"aiagentsdirectory","status":"pending","summary":"Wearable AI device that aims to replace smartphones","profileUrl":"https://agentstaffer.com/agents/aiagentsdirectory-humane-ai-pin","fitScore":44,"roleFamily":"Workflow Automation","reviewSignal":"Pending","urls":{"profileJson":"https://agentstaffer.com/agents/aiagentsdirectory-humane-ai-pin/profile.json","installJson":"https://agentstaffer.com/agents/aiagentsdirectory-humane-ai-pin/install.json","trustJson":"https://agentstaffer.com/agents/aiagentsdirectory-humane-ai-pin/trust.json","trialJson":"https://agentstaffer.com/agents/aiagentsdirectory-humane-ai-pin/trial.json"},"queueRank":7,"recommendation":{"label":"Needs review","score":44},"trialPriority":64,"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/aiagentsdirectory-humane-ai-pin#trial","runPath":"/agents/aiagentsdirectory-humane-ai-pin#trial","createEndpoint":"https://agentstaffer.com/api/actions/run-trial","createBody":{"listingId":"aiagentsdirectory:agent-humane-ai-pin","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/aiagentsdirectory-humane-ai-pin/profile.json","passed":true},{"id":"install","label":"Setup guide","href":"https://agentstaffer.com/agents/aiagentsdirectory-humane-ai-pin/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.","Stale source activity: Last visible update is about 774 days old."],"createRequest":{"endpoint":"https://agentstaffer.com/api/actions/run-trial","method":"POST","body":{"listingId":"aiagentsdirectory:agent-humane-ai-pin","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}}},"safety":{"trustTier":"Needs review","installReadiness":"Source guide","topRisk":{"level":"high","label":"Stale source activity","detail":"Last visible update is about 774 days old."},"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":"agentlocker:agent-humane-ai","slug":"agentlocker-humane-ai","name":"Humane AI","type":"agent","source":"agentlocker","status":"pending","summary":"The Humane AI Pin is a wearable AI-powered device designed to provide smartphone-like functionality without a","profileUrl":"https://agentstaffer.com/agents/agentlocker-humane-ai","fitScore":40,"roleFamily":"Coding & Development","reviewSignal":"Pending","urls":{"profileJson":"https://agentstaffer.com/agents/agentlocker-humane-ai/profile.json","installJson":"https://agentstaffer.com/agents/agentlocker-humane-ai/install.json","trustJson":"https://agentstaffer.com/agents/agentlocker-humane-ai/trust.json","trialJson":"https://agentstaffer.com/agents/agentlocker-humane-ai/trial.json"},"queueRank":8,"recommendation":{"label":"Needs review","score":40},"trialPriority":60,"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/agentlocker-humane-ai#trial","runPath":"/agents/agentlocker-humane-ai#trial","createEndpoint":"https://agentstaffer.com/api/actions/run-trial","createBody":{"listingId":"agentlocker:agent-humane-ai","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/agentlocker-humane-ai/profile.json","passed":true},{"id":"install","label":"Setup guide","href":"https://agentstaffer.com/agents/agentlocker-humane-ai/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":"agentlocker:agent-humane-ai","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":"claudemarketplaces:skills-orchestra-research-ai-research-skills-nemo-evaluator-sdk","slug":"claudemarketplaces-nemo-evaluator-sdk","name":"nemo-evaluator-sdk","type":"skill","source":"claudemarketplaces","status":"pending","summary":"NVIDIA's enterprise benchmarking platform that runs your LLMs through 100+ evaluation tasks from 18+ harnesses including MMLU, HumanEval, and GSM8K. Works with any OpenAI-compatible endpoint and handles execution across local Docker, Slurm HPC clusters, or cloud platforms. The containerized approach means reproducible results, and you get built-in exports t...","profileUrl":"https://agentstaffer.com/skills/claudemarketplaces-nemo-evaluator-sdk","fitScore":43,"roleFamily":"Coding & Development","reviewSignal":"Pending","urls":{"profileJson":"https://agentstaffer.com/skills/claudemarketplaces-nemo-evaluator-sdk/profile.json","installJson":"https://agentstaffer.com/skills/claudemarketplaces-nemo-evaluator-sdk/install.json","trustJson":"https://agentstaffer.com/skills/claudemarketplaces-nemo-evaluator-sdk/trust.json","trialJson":"https://agentstaffer.com/skills/claudemarketplaces-nemo-evaluator-sdk/trial.json"},"queueRank":9,"recommendation":{"label":"Needs review","score":43},"trialPriority":63,"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-nemo-evaluator-sdk#trial","runPath":"/skills/claudemarketplaces-nemo-evaluator-sdk#trial","createEndpoint":"https://agentstaffer.com/api/actions/run-trial","createBody":{"listingId":"claudemarketplaces:skills-orchestra-research-ai-research-skills-nemo-evaluator-sdk","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-nemo-evaluator-sdk/profile.json","passed":true},{"id":"install","label":"Setup guide","href":"https://agentstaffer.com/skills/claudemarketplaces-nemo-evaluator-sdk/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-orchestra-research-ai-research-skills-nemo-evaluator-sdk","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":"claudemarketplaces:skills-orchestra-research-ai-research-skills-evaluating-code-models","slug":"claudemarketplaces-evaluating-code-models","name":"evaluating-code-models","type":"skill","source":"claudemarketplaces","status":"pending","summary":"This wraps the BigCode Evaluation Harness, which benchmarks code generation models across 15+ standardized tests including HumanEval, MBPP, and MultiPL-E supporting 18 programming languages. You'd reach for this when you need to compare model performance systematically rather than relying on vibes, or when you're fine-tuning a code model and want real numbe...","profileUrl":"https://agentstaffer.com/skills/claudemarketplaces-evaluating-code-models","fitScore":36,"roleFamily":"Coding & Development","reviewSignal":"Pending","urls":{"profileJson":"https://agentstaffer.com/skills/claudemarketplaces-evaluating-code-models/profile.json","installJson":"https://agentstaffer.com/skills/claudemarketplaces-evaluating-code-models/install.json","trustJson":"https://agentstaffer.com/skills/claudemarketplaces-evaluating-code-models/trust.json","trialJson":"https://agentstaffer.com/skills/claudemarketplaces-evaluating-code-models/trial.json"},"queueRank":10,"recommendation":{"label":"Needs review","score":36},"trialPriority":56,"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-evaluating-code-models#trial","runPath":"/skills/claudemarketplaces-evaluating-code-models#trial","createEndpoint":"https://agentstaffer.com/api/actions/run-trial","createBody":{"listingId":"claudemarketplaces:skills-orchestra-research-ai-research-skills-evaluating-code-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/claudemarketplaces-evaluating-code-models/profile.json","passed":true},{"id":"install","label":"Setup guide","href":"https://agentstaffer.com/skills/claudemarketplaces-evaluating-code-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":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-orchestra-research-ai-research-skills-evaluating-code-models","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":"awesomeagentskills:davila7-claude-code-templates-tree-main-cli-tool-components-skills-ai-research-evaluation-bigcode-evaluation-harness","slug":"awesomeagentskills-evaluating-code-models","name":"evaluating-code-models","type":"skill","source":"awesomeagentskills","status":"pending","summary":"Evaluates code generation models across HumanEval, MBPP, MultiPL-E, and 15+ benchmarks with pass@k metrics. Use when benchmarking code models, comparing coding abilities, testing multi-language support, or measuring code generation quality. Industry standard from BigCode Project used by HuggingFace leaderboards.","profileUrl":"https://agentstaffer.com/skills/awesomeagentskills-evaluating-code-models","fitScore":36,"roleFamily":"Coding & Development","reviewSignal":"Pending","urls":{"profileJson":"https://agentstaffer.com/skills/awesomeagentskills-evaluating-code-models/profile.json","installJson":"https://agentstaffer.com/skills/awesomeagentskills-evaluating-code-models/install.json","trustJson":"https://agentstaffer.com/skills/awesomeagentskills-evaluating-code-models/trust.json","trialJson":"https://agentstaffer.com/skills/awesomeagentskills-evaluating-code-models/trial.json"},"queueRank":11,"recommendation":{"label":"Needs review","score":36},"trialPriority":56,"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/awesomeagentskills-evaluating-code-models#trial","runPath":"/skills/awesomeagentskills-evaluating-code-models#trial","createEndpoint":"https://agentstaffer.com/api/actions/run-trial","createBody":{"listingId":"awesomeagentskills:davila7-claude-code-templates-tree-main-cli-tool-components-skills-ai-research-evaluation-bigcode-evaluation-harness","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/awesomeagentskills-evaluating-code-models/profile.json","passed":true},{"id":"install","label":"Setup guide","href":"https://agentstaffer.com/skills/awesomeagentskills-evaluating-code-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":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":"awesomeagentskills:davila7-claude-code-templates-tree-main-cli-tool-components-skills-ai-research-evaluation-bigcode-evaluation-harness","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":"agencymarketplace:agents-humane","slug":"agencymarketplace-humane","name":"Humane","type":"agent","source":"agencymarketplace","status":"pending","summary":"See the world, not your screen.","profileUrl":"https://agentstaffer.com/agents/agencymarketplace-humane","fitScore":35,"roleFamily":"Workflow Automation","reviewSignal":"Pending","urls":{"profileJson":"https://agentstaffer.com/agents/agencymarketplace-humane/profile.json","installJson":"https://agentstaffer.com/agents/agencymarketplace-humane/install.json","trustJson":"https://agentstaffer.com/agents/agencymarketplace-humane/trust.json","trialJson":"https://agentstaffer.com/agents/agencymarketplace-humane/trial.json"},"queueRank":12,"recommendation":{"label":"Needs review","score":35},"trialPriority":55,"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/agencymarketplace-humane#trial","runPath":"/agents/agencymarketplace-humane#trial","createEndpoint":"https://agentstaffer.com/api/actions/run-trial","createBody":{"listingId":"agencymarketplace:agents-humane","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/agencymarketplace-humane/profile.json","passed":true},{"id":"install","label":"Setup guide","href":"https://agentstaffer.com/agents/agencymarketplace-humane/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":"agencymarketplace:agents-humane","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":"claudemarketplaces:skills-nohumanera-cn-stock-price-cn-stock-price","slug":"claudemarketplaces-cn-stock-price","name":"cn-stock-price","type":"skill","source":"claudemarketplaces","status":"pending","summary":"A straightforward Python script for checking real-time prices of Chinese A-share stocks. You run it from the command line either with default tickers (which includes major indices and some individual stocks) or pass in specific symbols like sh600519 or sz002352. It handles multiple stocks in one go, which is handy if you're monitoring a watchlist. The imple...","profileUrl":"https://agentstaffer.com/skills/claudemarketplaces-cn-stock-price","fitScore":40,"roleFamily":"Coding & Development","reviewSignal":"Pending","urls":{"profileJson":"https://agentstaffer.com/skills/claudemarketplaces-cn-stock-price/profile.json","installJson":"https://agentstaffer.com/skills/claudemarketplaces-cn-stock-price/install.json","trustJson":"https://agentstaffer.com/skills/claudemarketplaces-cn-stock-price/trust.json","trialJson":"https://agentstaffer.com/skills/claudemarketplaces-cn-stock-price/trial.json"},"queueRank":13,"recommendation":{"label":"Needs review","score":40},"trialPriority":60,"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-cn-stock-price#trial","runPath":"/skills/claudemarketplaces-cn-stock-price#trial","createEndpoint":"https://agentstaffer.com/api/actions/run-trial","createBody":{"listingId":"claudemarketplaces:skills-nohumanera-cn-stock-price-cn-stock-price","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":"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.","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/claudemarketplaces-cn-stock-price/profile.json","passed":true},{"id":"install","label":"Setup guide","href":"https://agentstaffer.com/skills/claudemarketplaces-cn-stock-price/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-nohumanera-cn-stock-price-cn-stock-price","reviewer":"Trial reviewer","notes":"Sandbox run using mock data."}}},"safety":{"trustTier":"Needs review","installReadiness":"CLI likely","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":"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":"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.","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":"remoteopenclaw:skills-hermes-built-in-evaluating-llms-harness","slug":"remoteopenclaw-evaluating-llms-harness","name":"Evaluating Llms Harness","type":"skill","source":"remoteopenclaw","status":"pending","summary":"Evaluates LLMs across 60+ academic benchmarks (MMLU, HumanEval, GSM8K, TruthfulQA, HellaSwag). Use when benchmarking model quality, comparing models, reporting academic results, or tracking training progress. Industry standard used by EleutherAI, HuggingFace, and major labs. Supports HuggingFace, vLLM, APIs.","profileUrl":"https://agentstaffer.com/skills/remoteopenclaw-evaluating-llms-harness","fitScore":38,"roleFamily":"Research & Knowledge","reviewSignal":"Pending","urls":{"profileJson":"https://agentstaffer.com/skills/remoteopenclaw-evaluating-llms-harness/profile.json","installJson":"https://agentstaffer.com/skills/remoteopenclaw-evaluating-llms-harness/install.json","trustJson":"https://agentstaffer.com/skills/remoteopenclaw-evaluating-llms-harness/trust.json","trialJson":"https://agentstaffer.com/skills/remoteopenclaw-evaluating-llms-harness/trial.json"},"queueRank":14,"recommendation":{"label":"Needs review","score":38},"trialPriority":58,"trial":{"title":"Run a cited research brief trial","input":"Ask for a one-page brief on a narrow question with source links and confidence notes.","expectedOutput":"A sourced summary that separates facts, assumptions, and unresolved questions.","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":["Cites primary sources","Flags uncertainty","Avoids unsupported claims","Keeps scope tight"],"runUrl":"https://agentstaffer.com/skills/remoteopenclaw-evaluating-llms-harness#trial","runPath":"/skills/remoteopenclaw-evaluating-llms-harness#trial","createEndpoint":"https://agentstaffer.com/api/actions/run-trial","createBody":{"listingId":"remoteopenclaw:skills-hermes-built-in-evaluating-llms-harness","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-evaluating-llms-harness/profile.json","passed":true},{"id":"install","label":"Setup guide","href":"https://agentstaffer.com/skills/remoteopenclaw-evaluating-llms-harness/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-hermes-built-in-evaluating-llms-harness","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":"deepnlpagents:theboringhumane-echoollama","slug":"deepnlpagents-echoollama-theboringhumane","name":"echoOLlama theboringhumane","type":"agent","source":"deepnlpagents","status":"pending","summary":"🦙 echoOLlama: A real-time voice AI platform powered by local LLMs. Features WebSocket streaming, voice interactions, and OpenAI API compa…","profileUrl":"https://agentstaffer.com/agents/deepnlpagents-echoollama-theboringhumane","fitScore":36,"roleFamily":"Workflow Automation","reviewSignal":"Pending","urls":{"profileJson":"https://agentstaffer.com/agents/deepnlpagents-echoollama-theboringhumane/profile.json","installJson":"https://agentstaffer.com/agents/deepnlpagents-echoollama-theboringhumane/install.json","trustJson":"https://agentstaffer.com/agents/deepnlpagents-echoollama-theboringhumane/trust.json","trialJson":"https://agentstaffer.com/agents/deepnlpagents-echoollama-theboringhumane/trial.json"},"queueRank":15,"recommendation":{"label":"Needs review","score":36},"trialPriority":56,"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/deepnlpagents-echoollama-theboringhumane#trial","runPath":"/agents/deepnlpagents-echoollama-theboringhumane#trial","createEndpoint":"https://agentstaffer.com/api/actions/run-trial","createBody":{"listingId":"deepnlpagents:theboringhumane-echoollama","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/deepnlpagents-echoollama-theboringhumane/profile.json","passed":true},{"id":"install","label":"Setup guide","href":"https://agentstaffer.com/agents/deepnlpagents-echoollama-theboringhumane/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":"deepnlpagents:theboringhumane-echoollama","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-vishisht16-2fhumane-proxy","slug":"mcpregistry-humane-proxy","name":"Humane Proxy","type":"tool","source":"mcpregistry","status":"pending","summary":"AI safety middleware — detects self-harm and criminal intent in LLM prompts.","profileUrl":"https://agentstaffer.com/mcp/mcpregistry-humane-proxy","fitScore":29,"roleFamily":"Workflow Automation","reviewSignal":"Pending","urls":{"profileJson":"https://agentstaffer.com/mcp/mcpregistry-humane-proxy/profile.json","installJson":"https://agentstaffer.com/mcp/mcpregistry-humane-proxy/install.json","trustJson":"https://agentstaffer.com/mcp/mcpregistry-humane-proxy/trust.json","trialJson":"https://agentstaffer.com/mcp/mcpregistry-humane-proxy/trial.json"},"queueRank":16,"recommendation":{"label":"Needs review","score":29},"trialPriority":49,"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-humane-proxy#trial","runPath":"/mcp/mcpregistry-humane-proxy#trial","createEndpoint":"https://agentstaffer.com/api/actions/run-trial","createBody":{"listingId":"mcpregistry:servers-io-github-vishisht16-2fhumane-proxy","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-humane-proxy/profile.json","passed":true},{"id":"install","label":"Setup guide","href":"https://agentstaffer.com/mcp/mcpregistry-humane-proxy/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-vishisht16-2fhumane-proxy","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."]}}}],"recentRuns":[],"pagination":{"page":1,"pageSize":24,"totalItems":16,"totalPages":1,"nextPageUrl":null,"prevPageUrl":null},"filters":{"q":"Humane","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":"cli","label":"CLI (7)"},{"value":"mcp","label":"MCP (7)"},{"value":"http","label":"HTTP (6)"},{"value":"sdk","label":"SDK (2)"}],"trust":[{"value":"needs-review","label":"Needs review (16)"}],"install":[{"value":"source-guide","label":"Source guide (8)"},{"value":"mcp-ready","label":"MCP ready (7)"},{"value":"cli-likely","label":"CLI likely (1)"}],"sources":[{"value":"claudemarketplaces","label":"claudemarketplaces (3)"},{"value":"mcpservers","label":"mcpservers (3)"},{"value":"playbooks","label":"playbooks (2)"},{"value":"agencymarketplace","label":"agencymarketplace (1)"},{"value":"agentlocker","label":"agentlocker (1)"},{"value":"aiagentsdirectory","label":"aiagentsdirectory (1)"},{"value":"awesomeagentskills","label":"awesomeagentskills (1)"},{"value":"deepnlpagents","label":"deepnlpagents (1)"},{"value":"explainx","label":"explainx (1)"},{"value":"mcpregistry","label":"mcpregistry (1)"},{"value":"remoteopenclaw","label":"remoteopenclaw (1)"}],"types":[{"value":"skill","label":"skill (10)"},{"value":"agent","label":"agent (5)"},{"value":"tool","label":"tool (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=Humane&sort=trial_readiness","note":"Action queue and page contract"},{"step":"03","label":"profile HTML","href":"/mcp/explainx-humane","note":"Human decision summary"},{"step":"04","label":"profile package","href":"/mcp/explainx-humane/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=Humane&sort=trial_readiness","trialsJson":"https://agentstaffer.com/trials.json?q=Humane&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"}}