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 skills add closedloop-ai/claude-plugins --skill gh-monitor-prgit clone --depth 1 https://github.com/closedloop-ai/claude-pluginsWrote 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/skills/closedloop-ai/claude-plugins/gh-monitor-pr)<a href="https://agentmods.dev/skills/closedloop-ai/claude-plugins/gh-monitor-pr"><img src="https://agentmods.dev/badge/skills/closedloop-ai/claude-plugins/gh-monitor-pr/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/skills/closedloop-ai/claude-plugins/gh-monitor-pr"><img src="https://agentmods.dev/badge/skills/closedloop-ai/claude-plugins/gh-monitor-pr.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 | $0.00081 | $0.04708 |
| Opus 5.5 | $0.00032 | $0.01883 |
| Sonnet 5.5 | $0.00016 | $0.00942 |
| Haiku 4.5 | $0.00008 | $0.00471 |
Grade A, and why
gh-monitor-pr 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 — 409 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Live-root validation boundary
Validation against a live root is limited to probe-only, read-only protocol calls. Never pipe synthetic events or user prompts into a live root. Run notification and event-delivery simulations only against an isolated fake server or a disposable, projectless test thread unambiguously owned by the test. Never fabricate text claiming that the user instructed a pause, stop, or mutation, and never stop or steer shared roots, workers, or the daemon. If isolated cleanup is unsupported, fail closed and report the limitation.
Monitor a GitHub PR
Start the detached monitor, report the handoff, and end the turn. Never poll, sleep, or use a tool-based wait in the launching Codex turn.
App Server wakeups use the native managed Codex App Server daemon through the
portable codex app-server proxy --sock transport. The monitor keeps its own
local delivery receipts: it persists a delivering attempt before invoking the
native notifier, and a restart never replays an unknown acceptance result.
Set up the App Server
Use the managed local App Server daemon. Never spawn a competing App Server and
never set CODEX_APP_SERVER_WS_URL.
For every command below, resolve skill_dir to the absolute directory
containing this SKILL.md; do not assume the skill lives in
${CODEX_HOME:-$HOME/.codex}/skills.
Treat Desktop and CLI setup as separate surfaces. If the launching root is a
CLI, use only the explicit cli setup below. Never run the no-argument Desktop
setup merely to support a CLI monitor: it changes the GUI launchctl
environment and installs a persistent Desktop launcher, neither of which is
needed for CLI monitoring. Do not install a periodic launcher that invokes
codex app-server daemon start for a CLI root. Start the managed daemon on
demand during CLI bootstrap instead.
For Codex Desktop, preserve the no-argument setup:
skill_dir='<absolute path to this gh-monitor-pr skill directory>'
node "${skill_dir}/scripts/setup-app-server.mjs"
What ships with it
13 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- agents/openai.yaml 257 B
- scripts/app-server-notify.js 752 B runs code
- scripts/app-server-notify.mjs 3.4 KB runs code
- scripts/app-server-notify.test.js 6.4 KB runs code
- scripts/app-server-start.zsh 652 B runs code
- scripts/client-process-api.mjs 5.3 KB runs code
- scripts/monitor-pr.mjs 117 KB runs code
- scripts/monitor-pr.test.mjs 62 KB runs code
- scripts/native-app-server-client.mjs 18 KB runs code
- scripts/process-identity.mjs 1.5 KB runs code
- scripts/setup-app-server.mjs 8.7 KB runs code
- scripts/setup-app-server.test.mjs 1.8 KB runs code
- scripts/skill-contract.test.mjs 763 B runs code
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 Changed · +24 lines c0b66ffa953a
- 4d ago First seen · 385 lines · 81 tokens per session scan A cf351434589b
gh-monitor-pr is a skill published in the GitHub repository closedloop-ai/claude-plugins (122 stars, last pushed today), licensed Apache-2.0. It adds 81 tokens to every session and 4,708 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-10-06.
Other skills, from other repositories
release-notes
Write release notes for a tagged release — a business-impact summary of what changed for the user, plus a thanks section naming every contributor whose commits are in the release. Short by construction.
autoship
Runs a changesets npm release through the version PR, CI publish, and registry verification. Use when asked to "release this package", "autoship", "merge Version Packages", or diagnose a release that did not publish.
pypi-release
This skill should be used when releasing tunacode-cli to PyPI. It keeps the existing local release checks, then hands the actual PyPI upload to a GitHub Actions workflow that uses the repository's PYPIAPITOKEN secret.
id-rollback
The infra-delivery INCIDENT safe-revert PROCEDURE skill (the recurring/incident rollback step) — the PROCEDURE the generic agent runs when a delivered change did NOT land (a NOT-LANDED realization, an escaped defect, bad config) to restore the last-known-good IaC and prove reality matches it again. The shape is git…
github
GitHub via gh CLI: PRs, issues, reviews, repos, auth.
changelog
Turn a range of commits or merged PRs into a changelog entry grouped by change type. Use when the user asks for release notes, a changelog, or "what changed" between two points.