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-reviewgit 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)<a href="https://agentmods.dev/skills/iusztinpaul/squid/squid-review"><img src="https://agentmods.dev/badge/skills/iusztinpaul/squid/squid-review/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"><img src="https://agentmods.dev/badge/skills/iusztinpaul/squid/squid-review.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 | $0.00054 | $0.00868 |
| Opus 5 | $0.00027 | $0.00434 |
| Sonnet 5 | $0.00011 | $0.00174 |
| Haiku 4.5 | $0.00005 | $0.00087 |
Grade A, and why
squid-review 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 3d 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 — 80 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Review — push, PA acceptance, PR review
Take the committed feature branch and turn it into a clean, pushed feature PR — or a rollup task describing what to fix. Both gates run on the pushed PR, sequentially: Product Architect (PA) acceptance first, then PR-Reviewer.
You are the orchestrator — a MANAGER. You push, launch the review agents, enforce the gates, and route failures back as rollup tasks. You do NOT review the diff, write code, or merge.
Read AGENTS.md first (tracker mode). Retry caps are below; the Severity Rule lives in the PR-Reviewer's role definition.
Input: a feature branch whose tasks are implemented and committed (the worktree when orchestrated by /squid-implement-night; the current branch when run standalone).
Output: a pushed feature PR with NO blockers (Nits appended to the PR description), OR ONE rollup task.
Critical rules:
- Never rubber-stamp an ACCEPT or NO BLOCKERS — spot-check the evidence; re-launch with feedback if it's thin.
- One rollup task per failed gate, never one ticket per issue.
Step 1 — Push + create/update the PR
Agent(
subagent_type="squid:software-engineer",
prompt="""Push feature {title} and open/update its PR per your Push / open PR section. {Working directory: {path}.}
The PR description summarizes the feature and lists each task by ID. Hand back the PR number."""
)
Capture the PR number for the gates below.
Step 2 — PA acceptance (Any product issues?)
Agent(
subagent_type="squid:product-architect",
prompt="""Acceptance review for feature {title} on PR #{N}. Read AGENTS.md first. Follow your acceptance-review role.
{Working directory: {path}.}
Walk the feature from the user's perspective against the Tasks Plan's acceptance criteria. Verdict: ACCEPT or REJECT. On REJECT, write ONE rollup task capturing ALL product issues."""
)
- ACCEPT (verified) → Step 3.
- REJECT → the PA filed ONE rollup task. Return it (see Output).
- Cap: PA REJECT max 3 per feature. On the 4th would-be REJECT, stop and surface
USER ACTION REQUIRED.
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.
- 3d ago Changed · -10 lines · -10 tokens per session 57ce01dee60c
- 9d ago First seen · 90 lines · 64 tokens per session scan A af06e8528640
squid-review is a skill published in the GitHub repository iusztinpaul/squid (186 stars, last pushed 5d ago), licensed Apache-2.0. It adds 54 tokens to every session and 868 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 2026-08-30.
Other skills, from other repositories
ai-assist-git-pr
Adaptive GitHub PR lifecycle skill — create PRs, write/update descriptions, investigate review comments (Copilot + human) with research and batch approval, and check merge readiness. Triggers on: create PR, open PR, describe PR, update description, PR body, check comments, copilot feedback, review comments, address…
pr-workflow
Pull request lifecycle: commit, codex review, sync, review, fix, status, cleanup, and PR mining. Use when user wants to commit changes, get a second-opinion code review from Codex, push changes, create a PR, check PR status, fix review comments, clean up branches after merge, or mine tribal knowledge from PR reviews.…
om-merge-buddy
Scan open pull requests via the configured tracker, classify merge readiness from labels, reviews, CI, and mergeability, then report which PRs can merge now and which ones are close but blocked.
om-auto-fix-pr
Drive an open PR to merge-ready from its number — merges the latest base, then loops review-autofix (om-auto-review-pr), built-in CI stabilization, and UI verification (om-auto-qa-pr) until approvable, green, and QA-evidenced. A --ci-only mode drives just CI green on a PR or a plain branch. Files follow-up issues for…
om-approve-merge-pr
Approve (submit an approving review) and squash-merge a PR given only its number, refusing when the QA gate or a blocking label forbids it. Routes fixable blockers to om-auto-fix-pr (red CI via its --ci-only mode, or conflicts and review problems via the full loop). Optionally file a follow-up issue at the same time.…
om-pr-autopilot
Diagnose what state one open PR is actually in — unfinished plan steps, missing review, unresolved conversations, red CI, base conflicts, missing QA evidence, merge-ready — then run the matching chain of om- skills in order and publish a status report. Use for "finish PR 123", "what is left on PR 123", "drive PR 123…