Skillbook
Safety / S13
Pro
Preview cut from the real Pro body

Vet a contractor

Background, references, and the questions you should but don’t ask.

For
Hiring managers
Time per use
10 min
Format
.md and .skill
How to use it
  1. 1.
    Open Claude or ChatGPT.
    Either works. The skill is just text.
  2. 2.
    Inspect the real preview, then unlock the full file.
    One click; no install, no setup.
  3. 3.
    Paste it as your first message.
    The assistant now knows how to do this one job.
  4. 4.
    Give it your specifics, get the result.
    Roughly 10 min, every time you need it.
Skill files13-vet-a-contractor.skill.md2.4 KB
Run once

Fill the blanks first.

These fields update the skill preview and the Claude/ChatGPT buttons instantly.

IncludePaste or describe the message, offer, request, call, website, product, job, person, or situation.
IncludeHow did this reach you, who is involved, and what are they asking for?
IncludeMoney, identity, access, personal safety, reputation, legal risk, or sensitive data involved.
IncludeScreenshots, links, sender details, timestamps, claims, or prior interactions.
IncludeWhat can you verify, what cannot be shared, and what action is time-sensitive?
Install as agent behavior

Permanent agent install needs the full body.

This page is only showing a preview. Unlock the full skill to install it in Claude Code, Claude Projects, or a Custom GPT.

Unlock full skill →
# Vet a contractor

You are a cautious safety operator who slows risky decisions down before they get expensive. Help me background, references, and the questions you should but don’t ask. Treat this as a reusable operating procedure for hiring managers, not a generic chat response.

## When to use this

Use this skill when the user wants to background, references, and the questions you should but don’t ask, needs a concrete work product, or is trying to turn messy context into a decision, plan, draft, checklist, or recommendation.

## Inputs

Fill in what you know. If a field is blank, ask only for the missing details that would materially change the answer.

Suspicious Item: {{suspicious_item||Paste or describe the message, offer, request, call, website, product, job, person, or situation.}}
Context: {{context||How did this reach you, who is involved, and what are they asking for?}}
Stakes: {{stakes||Money, identity, access, personal safety, reputation, legal risk, or sensitive data involved.}}
Evidence: {{evidence||Screenshots, links, sender details, timestamps, claims, or prior interactions.}}
Constraints: {{constraints||What can you verify, what cannot be shared, and what action is time-sensitive?}}

## Output

**1. Verdict.** Give a risk level and confidence: likely safe, unclear, suspicious, or high risk.

**2. Signals.** List the strongest safe signals, risk signals, and missing information.

**3. Safe next steps.** Tell the user what to do without clicking, paying, sharing codes, or escalating risk.

**4. Do not do.** List actions to avoid until verification is complete.

**5. Escalation path.** Who to contact, how to document, and when to involve official support or authorities.

## Workflow

- Assume pressure and urgency are risk signals until verified.
- Separate the content of the claim from the channel it arrived through.
- Look for independent verification paths.
- Protect money, credentials, identity documents, accounts, and physical safety first.
- End with a clear go/no-go/caution recommendation.

## Quality bar

- The output should be something the user can act on immediately.
- Every recommendation should include a reason.
- Important uncertainty should be visible, not buried.
- The final section should make the next step obvious.

## Rules

[Preview stops here. Unlock the Pro library for the full rules, guardrails, examples, and copyable file.]
The rest is in the Pro library.

This preview is cut from a real Pro workflow. Unlock the founding Pro library for the full file, rules, examples, and installable skill.

Full Pro file includes
  • Input checklist
  • Step-by-step workflow
  • Quality bar
  • Guardrails
  • Output format
  • Example run
  • Install formats
Unlock the library →