Files
skills/skills/kvb-skill-compile/SKILL.md
T
Kevin Veen-BirkenbachandClaude Opus 5.5 8b0c287ada fix(skills): trim 22 descriptions back under the listing budget
The skill listing stood at 1498 of 1500 words; the trimmed descriptions bring it to 1420, so the new kvb-check-ansible description fits without pushing others out.

🎯 Fix: 95 %
💥 Break: 1 %
💸 Cost: 99 %
⏱️ Time: 100 %
🛡️ Security: 100 %
🏆 Score: 95
🚦 Gate: ✅ 90/5, cost and time <= 100, security 100 %

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-07 10:47:08 +00:00

3.1 KiB

name, description
name description
kvb-skill-compile Turn deterministic skill steps (fixed command sequences) into tested scripts in the skill's own repo.

A deterministic step's result follows from its inputs alone: fixed command sequences, output parsing and filtering, counting, thresholds and arithmetic, format checks, sorting and lookups, state bookkeeping. Such steps become scripts and their prose shrinks to the call; choosing, explaining and asking stay prose. Nothing changes before the operator confirms.

Step 1: scope and owner

The skills named after the command, else every skill of the current repository. Per skill, its owning repository:

  • Skills repository: a first-party skill under its skills/, and every installed copy of one (~/.claude/skills, ~/.agents/skills, a project's rendered .claude/skills); edit the source, never the copy, by the kvb-skill-tune skill's Persistence section.
  • Another repository: a skill tracked in that repository's own tree (git ls-files lists it there), such as a project's .claude/skills.
  • Third party: a skills-lock.json entry from a foreign source is not compiled; propose the change upstream.

Step 2: inventory

Read-only. Per skill, every deterministic step with its evidence: how often sessions ran it (transcripts, rasci.py tools of the kvb-org-rasci register) and where an agent got it wrong. One line each: skill:step -> deterministic part -> inputs -> output -> runs per use.

Step 3: classify

Compile what runs often or went wrong; a one-off stays prose. Per candidate:

  • Language and form: the owning repository's own (Python standard library in the skills repository); no new language, framework or dependency without the operator.
  • Interface: every host path, name and value an explicit parameter (the kvb-code-clean skill), documented exit codes, --json when another script or skill consumes the output.
  • Location: skills/<skill>/scripts/<verb>.py with a test under tests/ in the skills repository; elsewhere the owning repository's convention, read from its AGENTS.md, CONTRIBUTING and Makefile first.

Step 4: listen and confirm

Run the kvb-think-listen skill on the open points. Then AskUserQuestion: one multiSelect with the candidates, nothing preselected, and per pick a single-select Write it / Revise it / Discard showing the drafted script and the prose diff.

Step 5: write

  • Write the script, then replace the compiled prose with its call (${CLAUDE_SKILL_DIR}/scripts/<verb>.py ...) and one line on reading its output.
  • Every script gets a unit test with a negative control shown once, run once watch.py admit --kind test-nostack (the kvb-org-organize skill) prints grant; make test runs through queue.py hold (kvb-org-queue).
  • In another repository follow its agent rules (claims, worktree, branch); a file another line owns goes to that line through the kvb-org-coordinate skill.
  • Leave the result uncommitted for the kvb-git-commit-next skill.

Output

| Skill | Step | Script | Repository | Prose lines cut | Verdict |

Steps left as prose stay with their reason.