evals
Skill24 starsChanged 42 days agoMITView on GitHub
Evaluate job-search pipeline code and outputs for correctness. Use when running code reviews, verifying pipeline outputs after a run, or checking ongoing pipeline health. Handles static analysis, runtime verification, and health monitoring.
7 other projects use a version of this.
Adopted by
Most recently taken up by
- scottlivingston/claude-skillstheir own version, first seen Nov 8, 2026
- renzrollon/interlocktheir own version, first seen Nov 5, 2026
- jstilb/ai-assistanttheir own version, first seen Nov 1, 2026
- rterakedis/leanwheel-skillstheir own version, first seen Oct 17, 2026
- Troy-LL/Unfurnishedtheir own version, first seen Sep 25, 2026
- johnviklund/agent-skillstheir own version, first seen Sep 12, 2026
Other versions7 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 7
- renzrollon/interlock★ 61same name
- johnviklund/agent-skills★ 2same name
- rterakedis/leanwheel-skills★ 2same name
- scottlivingston/claude-skills★ 1same name
- Troy-LL/Unfurnished★ 1same name
- jstilb/ai-assistant★ 0same name
- mj-deving/pai-skills★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository9 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- company-researchSkill
- cv-tailorSkill
- interview-prepSkill
- job-searchSkill
- job-trackerSkill
- onboardingSkill
- career-managerSkill
The file
We split this skill into the rules it gives an agent
2 rules. 0 are common, 1 is distinctive, and 1 is very specific to this project.
0
Common
Most of the 7 edited versions we read keep these.
1
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
- Distinctive: line 21
Run the relevant eval script for your situation
- Very specific: line 46
Does NOT evaluate: itself, other career-manager skills, or non-pipeline code
“Common” is measured: kept by at least 6 in 10 of the 7 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.
Files
Evals
Independent evaluation suite for the job-search pipeline. Lives outside the pipeline it evaluates.
Evaluation Levels
| Level | Script | When to Run |
| 0. Test Suite | evals/tests/ (pytest) | After any code change |
| 1. Code Review | evals/scripts/code_review.py | After building or changing pipeline code |
| 2. Runtime Verify | evals/scripts/runtime_verify.py | After a pipeline run completes |
| 3. Health Monitor | evals/scripts/health_monitor.py | Recurring — every pipeline run or 2-3x/week |
Process
- Run the relevant eval script for your situation
- Review the report output
- Fix issues (or flag for user approval if touching shared data)
- Re-run the eval to verify fixes
Usage
# Level 0: Full test suite
uv run python3 -m pytest evals/tests/ -v
# Level 1: Static code review of job-search pipeline
uv run evals/scripts/code_review.py
# Level 2: Validate outputs after a pipeline run
uv run evals/scripts/runtime_verify.py
# Level 3: Ongoing health monitoring
uv run evals/scripts/health_monitor.py
uv run evals/scripts/health_monitor.py --jsonScope
- Evaluates:
job-search/(scripts, data, caches) - Does NOT evaluate: itself, other career-manager skills, or non-pipeline code
- Also owns: the pytest suite in
evals/tests/andevals/schemas/eval-results.schema.json
Output Format
Each script prints a check-by-check report with pass/warn/fail status and returns non-zero exit code if critical checks fail.
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
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.