agentleFS
manuelschurrmanuelschurr/professional-twin/skills/review-job-offer/SKILL.md

review-job-offer

Skill0 starsChanged 7 months agoMITView on GitHub

Use when the user provides a job description (pasted text, file path, or URL) and wants to evaluate fit against their professional twin repository. Outputs a structured fit assessment, then optionally hands off to the tailor-application skill.

The file

We split this skill into the rules it gives an agent

12 rules. About 0 are common, 9 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈9

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Before doing anything else, verify the cwd is a twin repo.

    line 22
  2. Distinctive:

    If any of these are missing, bail out:

    line 28
  3. Distinctive:

    Read the job description: If the user pasted text, work with that.

    line 36
  4. Very specific:

    Read the full twin repo: Read every .md file in core-profile/, direction/, companies/, and application-materials/.

    line 37
  5. Very specific:

    Dealbreaker check: If ANY dealbreaker from direction/target-roles-and-criteria.md is hit, surface it immediately at the top of the response.

    line 38
  6. Distinctive:

    Compose the assessment: Use the format below.

    line 39
  7. Very specific:

    Offer the handoff: Ask the user if they want to draft tailored application materials.

    line 40
  8. Distinctive:

    Do not invent skills or experience the repo doesn't claim.

    line 75
  9. Distinctive:

    Do not be optimistic for the sake of optimism.

    line 76
  10. Distinctive:

    After presenting the assessment, ask:

    line 80
  11. Distinctive:

    If the user says yes:

    line 84
  12. Distinctive:

    If the user says no, end the session.

    line 89

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Review Job Offer

Overview

Given a job description, evaluate how well it fits the user's professional twin repository. Surface dealbreakers immediately. Then map matches, gaps, red flags, and give a clear recommendation.

This skill is read-only — it does not modify any file in the twin repo.

When to use

  • The user pastes a job description and asks "is this a fit?", "should I apply?", or similar.
  • The user provides a job description with no question — assume they want a fit assessment.
  • The user mentions a specific role they're considering and wants a structured evaluation.

Pre-flight check

Before doing anything else, verify the cwd is a twin repo. Check for:

  • A CLAUDE.md referencing professional-twin
  • A core-profile/ directory
  • A direction/ directory

If any of these are missing, bail out:

This doesn't look like a professional twin repository. cd into your twin repo before invoking review-job-offer, or run setup to create one.

Checklist

Complete in order. Do not skip steps.

  1. Read the job description. If the user pasted text, work with that. If they gave a file path, read it. If they gave a URL, ask the user to paste the content (do not try to fetch unless they explicitly tell you to).
  2. Read the full twin repo. Read every .md file in core-profile/, direction/, companies/, and application-materials/. You need the complete picture to evaluate fit — relevant evidence may live in unexpected files.
  3. Dealbreaker check. If ANY dealbreaker from direction/target-roles-and-criteria.md is hit, surface it immediately at the top of the response. Ask whether to continue with the rest of the evaluation. Do not bury this — dealbreakers are the most important output.
  4. Compose the assessment. Use the format below.
  5. Offer the handoff. Ask the user if they want to draft tailored application materials. If yes, dispatch tailor-application as a subagent with the JD and the fit assessment as context.

Output format

Structure the assessment in this order:

## Fit Assessment: [Role title at Company]

### Dealbreakers
> Either "None hit." or a clear list of which dealbreakers are triggered.

### Overall fit
> One of: **Strong fit** / **Moderate fit** / **Weak fit** / **Dealbreaker — do not apply**.
> One sentence justifying the verdict.

### What matches
> Bulleted list. Each bullet pairs a JD requirement with concrete evidence from the twin repo. Cite the source file.
> Example: "Building product orgs from scratch — Strategy Compass (`companies/strategy-compass.md`): built PM and design functions from zero, hired and coached the team."

### Gaps
> Bulleted list. For each gap, note how significant it is — is it disqualifying, is it learnable on the job, is it bridgeable from adjacent experience?

### Red flags or unknowns
> Anything in the JD that's a yellow flag — vague language, suspicious comp omissions, unclear reporting lines, anti-patterns the user has called out in their direction docs.

### Recommendation
> One of: **Apply**, **Skip**, **Apply if [condition]**, **Need more info before deciding**.
> One paragraph explaining.

Tone

  • Be direct and specific. Do not hedge.
  • Cite evidence — every match and gap should reference a file in the twin repo.
  • Do not invent skills or experience the repo doesn't claim. If the user might have something but it's not in the repo, flag it as "not captured yet — verify with user."
  • Do not be optimistic for the sake of optimism. The point of this skill is to save the user from wasting time on bad-fit applications.

Final step — handoff to tailor-application

After presenting the assessment, ask:

Want me to draft tailored application materials for this role?

If the user says yes:

  • Dispatch tailor-application as a subagent (use the Agent tool with subagent_type=general-purpose and prompt the subagent to use the tailor-application skill).
  • Pass the job description and the fit assessment as context.
  • Let the subagent run independently — it will return with drafted materials.

If the user says no, end the session.

What this skill does NOT do

  • It does NOT write any files in the twin repo. Read-only.
  • It does NOT draft application materials inline. That's tailor-application's job.
  • It does NOT fetch URLs unless the user explicitly tells you to.
  • It does NOT make up experience the user doesn't have.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.