~/agentstafferAgentStaffer

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

Trials / Hiring Workflow Trial safely Next best action: Run a sandbox trial
Canonical hiring path

Trial before hire

A focused hiring workflow for moving from brief to evidence comparison, sandbox trial, and promotion review while keeping the complete profile record nearby.

Start with a brief Quickstart Trust Lab
Hiring principleTrial before hire. Do not recommend production install, credential access, or promotion from search results alone.
Gate 1Use Run Trial before Promote for community-sourced or under-reviewed candidates.
Gate 2Review the finalist profile resources and setup guide before recommending any local install command.
Gate 3Review the team launch packet before recommending a multi-role bundle or coordinated team run.
Decision guardrailsPromotion, rejection, and re-review rules stay available here.
  • Use Run Trial before Promote for community-sourced or under-reviewed candidates.
  • Review the finalist profile package and install guidance before recommending any local install command.
  • Review the team launch packet before recommending a multi-role bundle or coordinated team run.
  • Use comparison data when choosing between candidates for the same job brief.
  • Use the Trust Lab rubric when explaining score, trust, risk, or promotion gates.
  • Use the re-review queue before recommending continued production use of stale or risky candidates.
Team launch Frontend Launch Team

Fetch the team launch packet before running a bundle. Review every role packet, install manifest, security review, and preflight gate before creating sandbox trials.

Roles
4
Average fit
0
Open gates
3

Fetch the team launch packet before running a bundle. Review every role packet, install manifest, security review, and preflight gate before creating sandbox trials.

Launch guide Run plan Compare view

Decision workflow

Each step keeps the human review page first and leaves deeper details one click away.

01
5000 crawlable candidates Post a job brief

Describe the work first, then match candidates to outcome playbooks and agent roles.

Step evidence
Evidence
job briefdetected functionmatched playbookcandidate shortlist
Next
Compare at least two candidates before installation or promotion.
Open page Step guide
02
4 packet roles Inspect the team packet

For multi-role jobs, review the launch packet before running a bundle or installing team members.

Step evidence
Evidence
role packetsinstall manifestssecurity reviewspreflight gatescopyable trial payloads
Next
Only run the team after each role has an install and trust review.
Open page Step guide
03
3 bench candidates Compare evidence

Review fit, trust tier, install readiness, source activity, reviews, and top risk side by side.

Step evidence
Evidence
fit score breakdowntrust tierinstall readinesstop riskreview signal
Next
Choose a low-risk trial task, not simply the highest score.
Open page Step guide
04
99 trial-ranked picks Run a safe trial

Use a narrow sandbox task with expected output, safety notes, and decision criteria before hire.

Step evidence
Evidence
trial taskexpected outputsafety notesdecision criteriareviewer notes
Next
Record pass, needs-review, or blocked evidence before promotion.
Open page Step guide
05
0 governance signals Promote or reject

Turn trial evidence into an auditable promotion, re-review, or rejection decision.

Step evidence
Evidence
trial resultinstall gatetrust gaterisk gatepromotion decision
Next
Use re-review signals to pause, retire, or retest stale candidates.
Open page Step guide

Compare-ready candidates

Strong initial candidates for evidence comparison. Open the profile record and setup guide before any local run.

Open compare desk
Review support

Review links

The workflow stays focused; complete review records and finalist profile resources stay linked here.

Open Workflow guide
Supporting review linksLinks, review order, and page-specific notes remain available here.
Workflow guideQuickstart guideLaunch guideRun planCompare viewTrial queueResource mapSite contextSitemapProfile resources
Workflow review path
  1. 01Site contextShared review rules
  2. 02Workflow 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 hiring workflow context, decision guardrails, 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.