- Home
- Skills
- Prompt & Skill Engineering
- Agentic Distiller
Works with the AI Agents you already use
Agentic Distiller
Audit a local repo's license, scaffold it into an Agent Skills package, and validate the packaging. Local Python, no API key.
$29
Agentic Distiller
Example session with this skill installed
Use agentic-distiller. Distill the local repo fixtures/sample-todo-cli into a skill named todo-cli-maintainer, running the license preflight first.
- Read your context and instructions
- Compiled the agentic distiller
Summary
The bundled fixture passed all three stages: license audit, scaffold, validation.
| Stage | Result |
|---|---|
| License audit | mit, verdict permissive, DISTILL |
| Scaffold | todo-cli-maintainer/ with SKILL.md, examples/, references/, scripts/, templates/ |
| Validation | PASS: frontmatter valid, 4 resource files, 1 Python script, 0 compile errors, 0 restricted-license hits, 0 warnings |
python3 skills/github-repo-skill-distiller/scripts/distill_local_repo.py \
--repo fixtures/sample-todo-cli \
--name todo-cli-maintainer \
--output /tmp/sku03-out
This is the output of the bundled fixture run, not a customer project.
Next steps
Open the generated SKILL.md and fill in the workflows. The scaffold is a skeleton by design. Re-run validate_distilled_skill.py after editing.
Connects securely to your tools. The creator never sees your data.
About this skill
Turn a local repository into an Agent Skills package, with a license check first
Point it at a local repo. It runs a deterministic license preflight, scaffolds a SKILL.md package, and validates the result. Standard-library Python 3.9+ only.
Who this is for: you use a local coding agent (Claude Code, Codex, Cursor, OpenCode) and want to turn a codebase or handbook into a reusable skill without pulling restricted-license code into a commercial package.
What you get
github-repo-skill-distiller:audit_repo_for_distillation.py(verdictsDISTILL,INTERNAL_ONLY,NEEDS_REVIEW),init_distilled_skill.py(scaffold),validate_distilled_skill.py(frontmatter, Python syntax, copyleft leaks), anddistill_local_repo.py(all three in one run).book-skill-distiller: templates and a rubric for distilling a manuscript or domain handbook into small operational skills.- A golden fixture (
fixtures/sample-todo-cli) that reproduces aPASSrun. INSTALL.md,FIXTURE.md,COMPATIBILITY.md, and a commercial single-user license.
Read before you buy
- Works on local paths. It does not clone remote repositories.
- Scaffolding produces a skeleton. You or your agent still fill in the workflows.
- The license verdict is a signal, not legal advice.
- Verified: the Python CLI on macOS arm64. Not verified: skill discovery in Claude Code, Codex, Cursor and OpenCode (see
COMPATIBILITY.md). - A free preflight-only version is on GitHub: qingsongcui/claude-code-skills.
Produces a skeleton; you or your agent fill in the workflows. Works on local paths only and does not clone remote repositories. Requires Python 3.9+. Skill discovery in Claude Code, Codex, Cursor and OpenCode is not verified.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 1 month ago
Requires