honest-tailor
CLAUDE.md0 starsChanged 29 days agoMITView on GitHub
A Claude Code plugin: skills tailor-cv and track-applications, scripts render_cv.py (Markdown CV to HTML/PDF/Word) and check_cv.py (traces CV claims to sources), and a benchmark of fictional fragments. No personal data in this repo, ever. It is public. All examples and test data use the fictional persona in benchmark/persona/ (Ofcom drama-range phone numbers, example.com addresses, Contoso/Fabrikam employers). The maintainer's real CV lives in a separate workspace outside this folder; never copy from it, quote it in tests, or name its employers,…
In this repository2 other files it gives its agents.
- tailor-cvSkill
- track-applicationsSkill
The file
Files
The Honest Tailor (honest-tailor): notes for Claude
A Claude Code plugin: skills tailor-cv and track-applications, scripts render_cv.py (Markdown CV to
HTML/PDF/Word) and check_cv.py (traces CV claims to sources), and a benchmark of fictional fragments.
The one rule
No personal data in this repo, ever. It is public. All examples and test data use the fictional persona in
benchmark/persona/ (Ofcom drama-range phone numbers, example.com addresses, Contoso/Fabrikam employers). The
maintainer's real CV lives in a separate workspace outside this folder; never copy from it, quote it in tests, or
name its employers, projects or applications here. dev/check_private_data.py enforces this (generic patterns plus
the gitignored .private-terms), through the pre-commit hook (git config core.hooksPath dev/hooks) and
tests/test_private_data.py.
Commands
- Tests:
python -m pytest - Benchmark:
python benchmark/run_benchmark.py(--writeupdatesbenchmark/RESULTS.md) - Leak check:
python dev/check_private_data.py(--stagedalso checks the commit author email) - Try the plugin:
claude --plugin-dir <path to this repo>, started in a CV workspace
Conventions
- Python 3.10+, stdlib only in
check_cv.py;render_cv.pyuses pypdf and python-docx optionally. - British English everywhere except the US region profile's own rules.
- Skill files use
${CLAUDE_PLUGIN_ROOT}for script paths. - A new benchmark category must be documented in
benchmark/README.md(a test enforces it). .gitignoredenies everything at the top level by default. A new top-level file or folder must be added to its allowlist on purpose; that is what keeps a CV workspace or export from being committed by accident.docs/readout-example.pngis a screenshot ofdocs/readout-example.html(headless Edge, 956x745). Edit the HTML, never the image, so the leak check scans the text behind it. Likewise.claude-plugin/icon.pngcomes fromdocs/icon-source.html(1024x1024). The directory takes a plugin's icon only the first time it is saved there.- These notes live in
.claude/CLAUDE.md, not the repo root: a root CLAUDE.md is flagged by the plugin directory.
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.

