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.
git clone --depth 1 https://github.com/kLOsk/adloopWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/rules/klosk/adloop/releasing)<a href="https://agentmods.dev/rules/klosk/adloop/releasing"><img src="https://agentmods.dev/badge/rules/klosk/adloop/releasing/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/rules/klosk/adloop/releasing"><img src="https://agentmods.dev/badge/rules/klosk/adloop/releasing.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What 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 5.1 | $0.00000 | $0.01222 |
| Opus 5.5 | $0.00000 | $0.00489 |
| Sonnet 5.5 | $0.00000 | $0.00244 |
| Haiku 4.5 | $0.00000 | $0.00122 |
Grade A, and why
releasing 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 1mo ago.
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 — 76 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Releasing AdLoop
How publishing works
AdLoop uses PyPI Trusted Publishing via GitHub Actions. No API tokens or secrets are stored in the repo. PyPI verifies the publish request came from this exact repo and workflow via OIDC.
- CI (
.github/workflows/ci.yml): runs tests on every push tomainand every PR, across Python 3.11-3.13. - Release (
.github/workflows/release.yml): triggered by pushing av*tag. Builds the package, publishes to PyPI, and creates a GitHub Release.
When to release
Release after merging PRs that add features, fix bugs, or change the public tool interface. Not every merge needs a release — batch small fixes.
Release process
When the user asks to release, or after significant changes are merged:
-
Decide the version bump following semver:
- Patch (0.2.1): bug fixes, doc changes, internal refactors
- Minor (0.3.0): new tools, new parameters on existing tools, new features
- Major (1.0.0): breaking changes to tool signatures or config format
-
Bump the version in both places (they must match):
pyproject.toml→version = "X.Y.Z"src/adloop/__init__.py→__version__ = "X.Y.Z"
-
Commit and tag:
git add pyproject.toml src/adloop/__init__.py git commit -m "Release vX.Y.Z: <one-line headline>" git tag vX.Y.Z git push && git push --tags -
GitHub Actions handles the build and publish: tests run, package builds, publishes to PyPI, creates a GitHub Release. The auto-created release body is just a compare link — the human-written notes go on next (step 5).
-
Write the GitHub Release notes — this is the canonical changelog. We do not maintain a
CHANGELOG.md; every version's notes live on its GitHub Release page (e.g. https://github.com/kLOsk/adloop/releases/tag/v0.6.0). Draft the notes locally, then attach them once the release workflow has finished creating the release:gh release edit vX.Y.Z --notes-file /tmp/adloop-vX.Y.Z-notes.md
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.
- 1mo ago First seen · 76 lines · 0 tokens per session scan A 214886999edb
releasing is a cursor rule published in the GitHub repository kLOsk/adloop (282 stars, last pushed yesterday), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,222 tokens. 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 2026-08-30.
Other cursor rules, from other repositories
release
Changesets release pipeline - version PR and its dispatched CI check, shared vX.Y.Z tag, CI gate before tagging, release:check gate, Dependabot changesets, no npm publish, GITHUBTOKEN tag trigger limitation.
changesets
Writing changesets - when one is needed, apps-only frontmatter, bump levels for deployed apps, summary style, non-interactive creation.
cd
GitHub Actions CD - called by release.yml after the tag, no push-tags trigger, cf workers versions create then a 100% deployment in deployOrder, scoped API token on the production environment, GitHub-owned deployment record.
zip-contents-protection.crux
⟦CRUX:zip-contents-protection.md P.⊛{¬auto-Δ scripts/create-crux-zip.py w/o user req!}.
version-bump.crux
⟦CRUX:version-bump.md Κ{M=X.0.0;m=0.X.0;p=0.0.X} R{fix|docs|style|refactor|perf|test|chore→p;feat→m;BREAKING|!→M} Λ{⊲commit→Δ.crux/crux.json»stage} P{1bump;↑→↓0;¬conv→?} ⟧.
roslyn-mcp-github-release
GitHub Releases are dispatched by a human; agents never push tags.