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 iusztinpaul/squid --skill squid-review-cigit clone --depth 1 https://github.com/iusztinpaul/squidWrote 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/iusztinpaul/squid/squid-review-ci)<a href="https://agentmods.dev/skills/iusztinpaul/squid/squid-review-ci"><img src="https://agentmods.dev/badge/skills/iusztinpaul/squid/squid-review-ci/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/iusztinpaul/squid/squid-review-ci"><img src="https://agentmods.dev/badge/skills/iusztinpaul/squid/squid-review-ci.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00047 | $0.00633 |
| Opus 5 | $0.00023 | $0.00316 |
| Sonnet 5 | $0.00009 | $0.00127 |
| Haiku 4.5 | $0.00005 | $0.00063 |
Grade A, and why
squid-review-ci 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 7d 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 — 57 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Review CI — validate the pushed PR's pipeline
Watch CI for the pushed feature PR and drive it to green. On failure, the On-Call engineer diagnoses and hands a fix task to the SWE (it does not fix the code itself); once CI is green, the PR is ready to merge.
You are the orchestrator — a MANAGER. You launch On-Call and the SWE, enforce the cap, and confirm green. You do NOT read the diff for review (that was /squid-review).
Read AGENTS.md first.
Input: a pushed, review-clean feature PR.
Output: a CI-validated "ready to merge" PR (or USER ACTION REQUIRED if CI can't be made green within the cap).
The On-Call → SWE → re-check loop
Watch CI (On-Call)
Agent(
subagent_type="squid:oncall-engineer",
prompt="""Watch CI for the latest push on feat/{slug} (PR #{N}). {Working directory: {path}.} Read AGENTS.md first. Follow your role definition. CI/CD only — do not review the diff. Report green, or hand back a concrete FIX TASK (root cause + the exact failing command + affected files)."""
)
- Green → done. The PR is validated and ready to merge.
- Red (infra / flake) → On-Call files a NEW infra task and reports it; surface to the human — it does not count against the cap.
- Red (real failure) → On-Call hands back a fix task → go to "SWE fixes".
SWE fixes
Agent(
subagent_type="squid:software-engineer",
prompt="""CI failed on PR #{N}. {Working directory: {path}.} Fix task from On-Call: {root cause + failing command + files}. Reproduce locally with the same command CI ran, fix the root cause (not the symptom; fix the bug, not the test), re-run the local suite, commit per your Commit section (`Refs #N` — the original task is already closed), and push."""
)
Then re-run Watch CI (On-Call) to confirm the new run goes green.
Cap: 5 fix cycles per PR
If CI is still red after 5 On-Call→SWE→re-check cycles, stop, surface USER ACTION REQUIRED with the last failing log, and end.
Output
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.
- 7d ago Changed · -10 lines · -11 tokens per session 53848ae020f1
- 12d ago First seen · 67 lines · 58 tokens per session scan A d8c092f0255d
squid-review-ci is a skill published in the GitHub repository iusztinpaul/squid (190 stars, last pushed 8d ago), licensed Apache-2.0. It adds 47 tokens to every session and 633 once invoked, about $0.0002 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 2026-08-30.
Other skills, from other repositories
gh-fix-ci
Use when a user asks to debug or fix failing GitHub PR checks that run in GitHub Actions; use gh to inspect checks and logs, summarize failure context, draft a fix plan, and implement only after explicit approval. Treat external providers (for example Buildkite) as out of scope and report only the details URL.
git-ci
CI/CD status queries for GitHub Actions (gh) and GitLab CI (glab). Check pipeline status, failing jobs, workflow runs, job logs, and merge readiness. Use when checking CI status, debugging build failures, viewing job logs, watching pipeline progress, or configuring gh/glab CI read-only allowlists. Not for PR workflows…
ci-fix
Diagnose and fix a failing CI / GitHub Actions run. Use when a pipeline or build breaks. Triggers on: ci quebrou, ci falhou, conserta o ci, arruma o ci, build quebrou, o actions ta vermelho, pipeline failing, workflow failed, fix ci.
troubleshooting-whats-not-working
Use when something is broken, failing, or not behaving as expected (any domain - software, gadget, process, plan).
repo-hygiene
Periodic repository health check — dependencies, git, CI/CD, code quality, docs, security. Use when: onboarding to a repo, weekly maintenance, after big refactors, before audits, "is this repo in good shape?". Triggers: "repo hygiene", "health check", "repo health", "clean up repo", "maintenance check", "audit repo"…
gradle-build-performance
Debug and optimize Android/Gradle build performance. Use when builds are slow, investigating CI/CD performance, analyzing build scans, or identifying compilation bottlenecks.