Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/hautc-it/cil/releasegit clone --depth 1 https://github.com/hautc-it/cilWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00060 | $0.02491 |
| Opus 5 | $0.00030 | $0.01246 |
| Sonnet 5 | $0.00012 | $0.00498 |
| Haiku 4.5 | $0.00006 | $0.00249 |
Grade A, and why
release scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 269 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/release
Bump version, refresh CHANGELOG + README, stage commit/tag, hand off manual npm publish to the user.
Argument $ARGUMENTS is one of:
patch/minor/major— semver bump from currentpackage.jsonversionx.y.z— explicit version (e.g.1.2.0)- empty — ask user
Stop and ask the user at every checkpoint marked 🛑. Never run npm publish — that is the user's call. git push and PR creation are done by /release itself, but only after explicit user confirmation at the corresponding 🛑.
Step 1 — Pre-flight
Run in parallel:
git status --porcelain
git rev-parse --abbrev-ref HEAD
node -p "require('./package.json').version"
git describe --tags --abbrev=0 2>&1 || echo "NO_TAG"
Block release if:
- working tree dirty → ask user to commit/stash first
- branch ≠
main→ confirm with user before proceeding - last tag missing → treat all commits as the changelog window
🛑 Verify with user: current version X.Y.Z, last tag vX.Y.Z, branch main, tree clean.
Step 2 — Decide new version
$ARGUMENTS |
Action |
|---|---|
patch |
bump Z (e.g. 1.1.0 → 1.1.1) |
minor |
bump Y, reset Z (e.g. 1.1.0 → 1.2.0) |
major |
bump X, reset Y/Z (e.g. 1.1.0 → 2.0.0) |
x.y.z |
use exactly |
| empty | inspect commits (Step 3) and recommend |
Recommendation rule (when arg is empty):
- any
feat:commit →minor - only
fix:/chore:/docs:/refactor:→patch - any
BREAKING CHANGE:footer orfeat!:→major
🛑 Confirm new version with user before continuing.
Step 3 — Collect commits since last tag
git log <last-tag>..HEAD --pretty=format:"%h %s"
If empty → bail: "Nothing to release since <last-tag>."
Classify each commit by conventional prefix into buckets:
| Prefix | CHANGELOG section |
|---|---|
feat: / feat(...): |
Added |
fix: / fix(...): |
Fixed |
refactor: / perf: |
Changed |
revert: / removal feat |
Removed |
docs: / chore: / test: / ci: / build: |
skip (unless user-visible) |
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 269 lines · 60 tokens per session scan A e9e62869622f
release is a command published in the GitHub repository hautc-it/cil (1 stars, last pushed 1mo ago), licensed MIT. It adds 60 tokens to every session and 2,491 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
README
Git workflow and quality assurance commands for the claude-skills repository.
validate
Validate a finding — runs 7-Question Gate + 4-gate checklist. Kills weak findings before report writing. Prevents N/A submissions that hurt validity ratio. Usage: /validate.
convert-to-todowrite-tasklist-prompt
Purpose: Transform verbose, context-heavy slash commands into efficient TodoWrite tasklist-based methods with parallel subagent execution for 60-70% speed improvements.
security-audit
Perform a comprehensive security audit of the codebase to identify potential vulnerabilities, insecure patterns, and security best practice violations.
statusbar-style
Switch the status-bar style (classic / capsule / hairline).
session-end
I'll summarize this coding session and update the memory system with our accomplishments.