agentleFS
Sign inStart free
tydev-newtydev-new/10xjobs-careercoach/skills/evaluate/SKILL.md

evaluate

Skill3 starsChanged 52 days agoMITView on GitHub

Use this skill when the candidate wants to evaluate a role, a job description, or a company — e.g. "should I apply to this?", "evaluate this role at Writer", "decode this JD", "research this company", "compare these three roles". Produces one combined verdict from JD decoding + company research, persists the analyses, and records the verdict in the pipeline record (jobs.md).

25 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use compare, triage and 5 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • n24q02mSkill

    compare

    Structured comparison of 2+ alternatives with consistent criteria and decision matrix

    Also in 3 projects that use this

  • Google CloudSkill

    triage

    Gather context on a change or bug and plan the implementation.

    Also in 4 projects that use this

  • BlockSkill

    plan

    Draft a TDD, PRD, or lightweight plan in the conversation without committing it.

    Also in 3 projects that use this

  • SentrySkill

    review

    Fixture review skill.

    Also in 4 projects that use this

  • NVIDIASkill

    research

    Before designing a feature, research how comparable libraries solve the problem, check best practices, known pitfalls, and upstream changes. Use when starting a new feature, investigating an unfamiliar area, or the user asks to research something.

    Also in 3 projects that use this

  • VercelSkill

    setup

    Set up Vercel CLI and project configuration. Use when the user says "set up Vercel", "configure Vercel", "link to Vercel", or "vercel init".

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

3 rules. 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 24 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Exits: when batch is scored and candidate selects roles for full graduation.

    line 53
  2. Very specific:

    Session close: Run ../profile/scripts/check_files.mjs --workspace ..

    line 60
  3. Distinctive:

    Never synthesize uncertain sources into confident claims.

    line 65

“Common” is measured: kept by at least 6 in 10 of the 24 edited versions of this file we could read. “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

Evaluate — role + company, one verdict

Goal

One decision-ready verdict per role — JD decoding combined with company research into an actionable tier. The candidate decides whether to apply; you provide the odds. Scored by references/eval.md.

Must be trueWhere
Verdict recorded in jobs.md via script with exact company + title stringsjobs.md
Track assigned first and named in verdict, row, and analysis headerall three
Hard dealbreakers dismissed immediately with quoted reason before research spendjobs.md (dq: <reason>)
JD decode on file (lenses, confidence labels, competencies, 5-dim fit)jd-analysis/<key>.md
Company brief on file (tiered claims, dated, knowledge gaps noted)company/<slug>.md
Cheap tier stamped quick-scan: and graduations note prior scorejobs.md reason · chat
Summary card delivered in replyreply

Prerequisites

  • Required: Workspace CLAUDE.md (none → profile's Setup); JD or company name.
  • Optional: profile.md + criteria.md (grounds fit scoring), storybank.md (competency evidence), company/<slug>.md (cached research).

Sequences

Decoding lenses, fit algebra, and company research craft live in references/patterns.md. Verdict tiers in references/eval.md.

Full evaluation (a sequence)

Runs when candidate asks about a role or company, or other skills route for a missing analysis.

  1. Near-match check: Match against existing jobs.md rows before creating duplicates.
  2. Dealbreaker gate: Evaluate against criteria.md § Dealbreakers / § Compensation / § Geo. Hard hit $\rightarrow$ record the DQ immediately (no row yet → create it first; exact commands in references/patterns.md step 2, "Dealbreakers") and stop before research spend.
  3. Track assignment: Assign track from criteria.md § Targets before judging.
  4. Company research & Decode: Research company signals and decode JD across the 6 lenses (references/patterns.md).
  5. Verdict & Record: Combine into final verdict tier — Strong Fit (strong) / Investable Stretch (investable_stretch) / Long-Shot Stretch (long_shot) / Weak Fit (weak). Record via scripts/record_verdict.mjs --verdict <value>.
  6. Deliver summary card: Present the evaluation card in chat (references/schema.md § The summary card).

Exits with jobs.md row updated, analyses persisted to disk, and summary card delivered.

Quick-scan tier (a sequence)

Runs when high volume calls for a lightweight screening pass (batch triage, pipeline over cap).

  1. Run Dealbreaker gate.
  2. Decode fit against criteria.md + profile.md only (skip company research file and full jd-analysis file).
  3. Record verdict with quick-scan: prefix in jobs.md Reason field.
  4. Graduate selected roles to full evaluation before human investment, noting previous score.

Exits when batch is scored and candidate selects roles for full graduation.

State

Owned: jd-analysis/ and company/ (shapes in references/schema.md).

  • Hands back: Strong verdict → apply; interview booked → interview; competency extractions → storybank.
  • Session close: Run ../profile/scripts/check_files.mjs --workspace .. No checker-subagent is spawned (analyst voice). Report outcomes, never narration (clean is 1 line; fix FAILs before reply ends).

Guardrails

  • Honest verdicts based on evidence, never cheerleading or artificial pessimism.
  • Never synthesize uncertain sources into confident claims; conflicting data is explicitly presented.

Every reply ends with ONE contextual next step — a sentence with its why, not a menu.

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.

Check your setup, free