~/agentstafferAgentStaffer

Find, trial, compare, and govern AI agents before hiring

Trials / Terminal Trial safely Next best action: Run a sandbox trial
Agent hiring terminal

Operate AgentStaffer like a hiring CLI

A practical command workspace for finding candidates, checking profile evidence, and running sandbox-first AgentStaffer actions with endpoint details folded nearby.

Quickstart Trial queue API reference
Safety rule

Never install or hire directly from search results. Check profile evidence and setup guidance, run a sandbox trial, compare evidence, then promote or retire with governance signals.

Command paths

4 primary routes are visible; deeper lifecycle routes stay folded with examples and requests intact.

searchagentstaffer search
Find candidates by job, protocol, role, source, trust, or install readiness.
Command snippetUse after route selection 
agentstaffer search --brief 'Need a Next.js accessibility reviewer' --sort trial_readiness
Request exampleOptional API request 
curl -s 'https://agentstaffer.com/job-matcher.json?brief=Need%20a%20Next.js%20accessibility%20reviewer' | jq '.candidates[0:5]'

Open the top candidate profile evidence and compare before install.

Open page Command guide
profileagentstaffer profile
Read a candidate package dossier before executing anything locally.
Command snippetUse after route selection 
agentstaffer profile npm-iflow-mcp-sixhq-overture
Request exampleOptional API request 
curl -s 'https://agentstaffer.com/skills/npm-iflow-mcp-sixhq-overture/profile.json' | jq '{name:.agent.name, summary:.agentSummary, lifecycle:.hiringLifecycle, governance:.lifecycleGovernance}'

Check lifecycle governance and risk signals before trial or install.

Open page Command guide
installagentstaffer install
Fetch the setup guide without hiding risk or source evidence.
Command snippetUse after route selection 
agentstaffer install npm-iflow-mcp-sixhq-overture --dry-run
Request exampleOptional API request 
curl -s 'https://agentstaffer.com/skills/npm-iflow-mcp-sixhq-overture/install.json' | jq '.install.quickInstall'

Run only in a sandbox workspace until a trial is recorded.

Open page Command guide
trialagentstaffer trial
Create or inspect sandbox-first trial evidence before hiring.
Command snippetUse after route selection 
agentstaffer trial npm-iflow-mcp-sixhq-overture --reviewer 'Trial reviewer'
Request exampleOptional API request 
curl -s 'https://agentstaffer.com/trials.json?sort=trial_readiness&pageSize=12' | jq '.trialQueue[0]'

Store output, logs, safety notes, and reviewer decision criteria.

Open page Command guide
Show 3 more command pathsAdditional command paths stay folded for scanning; examples and request snippets remain available here.
compareagentstaffer compare
Compare candidates by fit, trust, install, source activity, risk, and first trial.
Command snippetUse after route selection 
agentstaffer compare npm-iflow-mcp-sixhq-overture npm-overture-mcp npm-awaithumans
Request exampleOptional API request 
curl -s 'https://agentstaffer.com/compare.json?ids=npm-iflow-mcp-sixhq-overture%2Cnpm-overture-mcp%2Cnpm-awaithumans' | jq '.candidates[] | {name, fitScore, trustTier, installReadiness}'

Pick the safest trial candidate, not merely the highest score.

Open page Command guide
promoteagentstaffer promote
Move tested candidates toward shortlist, publication, or active Office work after evidence review.
Command snippetUse after route selection 
agentstaffer promote --from /promotions.json
Request exampleOptional API request 
curl -s 'https://agentstaffer.com/promotions.json' | jq '.promotionQueue[0:5] | map({agent:.agent.name, action:.promotionDecision.nextAction, gates:.governance.gatesPassed})'

Do not promote if install, reputation, trial, or high-risk gates are not clear.

Open page Command guide
retireagentstaffer retire
Find stale, failed, risky, or low-evidence agents that need re-review or retirement.
Command snippetUse after route selection 
agentstaffer retire --queue /re-review.json
Request exampleOptional API request 
curl -s 'https://agentstaffer.com/re-review.json' | jq '.reReviewQueue[0:5] | map({agent:.agent.name, action:.recommendedAction, priority:.priorityScore})'

Pause before production use when risk, stale source activity, or failed trials appear.

Open page Command guide

Candidate shortcuts

3 candidates are visible for quick triage; the remaining shortcuts stay folded with review records intact.

Open trial queue
Show 2 more candidate shortcutsAdditional shortcuts stay folded for scanning; profiles, setup guides, and review links remain available here.
Review support

Review links

Command-style UI stays compact; complete review records and finalist profile resources stay linked here.

Open Command guide
Supporting review linksLinks, review order, and page-specific notes remain available here.
Command guideWorkflow guideQuickstart guideFind candidates by job, protocol, role, source, trust, or install readiness.Read a candidate package dossier before executing anything locally.Fetch the setup guide without hiding risk or source evidence.Create or inspect sandbox-first trial evidence before hiring.Compare candidates by fit, trust, install, source activity, risk, and first trial.Move tested candidates toward shortlist, publication, or active Office work after evidence review.Find stale, failed, risky, or low-evidence agents that need re-review or retirement.Trial queuePromotion queueRe-review queueResource mapSite contextSitemapProfile resources
Terminal review path
  1. 01Site contextShared review rules
  2. 02Command guideOpen the focused review record
  3. 03Profile pageOpen the candidate page
  4. 04Profile resourcesEvidence, setup, trust, trial, and history files

Use the links and review path below for command guidance, sandbox execution context, and finalist profile resources before action.

Review route: OpenAI review tools can start with openai.json; otherwise use site context and agent inventory before opening the finalist's complete review materials before install, hire, or promotion.