Skip to score review
~/agentstafferAgentStaffer

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

Insights / Scores Review evidence health Next best action: Open the highest-priority queue
Recommendation Scores

Scores with evidence attached

Public recommendation score center for explaining AgentStaffer recommended scores, score bands, evidence gaps, and the next safe action before an AI agent is installed or hired.

Score guideEvidence ledgerScoring rubric

Score overview

Start with the recommendation mix; open band detail when you need full score ranges.

Score method
Public candidates5000Reviewable inventory
Recommended0Evidence-adjusted shortlist
Needs evidence4715Review before ranking
Ledger entries0Score evidence
Recommendation score bandsBand ranges, inventory share, and next action stay available here.
Recommended0

88-100 score · 0% of public inventory.

Compare, then run a sandbox trial before promotion.
Worth trialing282

72-87 score · 6% of public inventory.

Run trial and collect evidence before install.
Needs evidence4715

50-71 score · 94% of public inventory.

Collect source, install, review, or trial evidence.
Review first3

0-49 score · 0% of public inventory.

Do not recommend without manual review.
Scoring Dimensions

What moves the score

Dimension 1Fitbreakdown.fit and breakdown.marketFit
Dimension 2Community reviewsbreakdown.community and rationale
Full method6 dimensionsEvidence-adjusted scoring
Dimension detailScoring dimensions, meanings, and citation paths stay available here.
  • FitSource relevance, market fit, capability metadata, and protocol compatibility.breakdown.fit and breakdown.marketFit
  • Community reviewsDirect AgentStaffer user evaluation signal. Missing reviews lower confidence.breakdown.community and rationale
  • Third-party reputationCrawled third-party review or discussion signal. Missing third-party reputation evidence is visible.breakdown.external and evidenceLedger
  • Sandbox trialsObserved trial runs or the lack of them. Trials are required before promotion.breakdown.trial and /trials.json
  • Install readinessWhether a candidate has MCP, CLI, source guide, or manual install state.breakdown.install and installJson
  • Risk penaltySecurity, staleness, low-evidence, or governance concerns subtracted from the recommendation.breakdown.riskPenalty and /trust-lab/rubric.json
Recent Evidence

Evidence that affects recommendation

LedgerNo score evidence yetReviews, trials, and crawled reputation signals will appear here.
Evidence detailRecent evidence records, scores, types, next actions, and summaries stay available here.

Top recommended summary

A short score-methodology sample. Use the full score guide for the complete ranked dataset and fetch profile evidence before action.

Score overview
Show 3 more top recommended candidatesAdditional score samples stay folded for scanning; scores.json and card review links keep the full evidence path.
Worth testing security-review
81
Score detailRationale and fit metrics stay available here.

Worth testing at 81/100 · MCP ready install readiness · High traction trust tier · no review or external signal yet

Fit
81 / 100
Action
collect evidence
Market Fit
144 / 100
Profile
Review links
Profile JSON Score JSON Evidence JSON Setup JSON

High fit, missing evidence

A short review sample for promising candidates that still need reviews, external reputation, or sandbox trial evidence.

Evidence queue
Show 3 more high-fit candidatesAdditional score samples stay folded for scanning; scores.json and card review links keep the full evidence path.
Worth testing security-review
81
Score detailRationale and fit metrics stay available here.

Worth testing at 81/100 · MCP ready install readiness · High traction trust tier · no review or external signal yet

Fit
81 / 100
Action
collect evidence
Market Fit
144 / 100
Profile
Review links
Profile JSON Score JSON Evidence JSON Setup JSON
Review support

Review links

Score guidance stays readable; evidence queues and profile resource order stay linked here.

Open Score guide JSON
Supporting review linksLinks, review order, and page-specific notes remain available here.
Score guide JSONEvidence ledger JSONEvidence queue JSONResource catalog JSONReview guide JSONAgent inventory JSONSite context TXTSitemap XML
Score review path
  1. 01Site contextShared review rules
  2. 02Score centerOpen 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 complete score guidance, evidence queues, 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.