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/MadaraUchiha-314/the-loopWrote 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/commands/madarauchiha-314/the-loop/review-pr)<a href="https://agentmods.dev/commands/madarauchiha-314/the-loop/review-pr"><img src="https://agentmods.dev/badge/commands/madarauchiha-314/the-loop/review-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/commands/madarauchiha-314/the-loop/review-pr"><img src="https://agentmods.dev/badge/commands/madarauchiha-314/the-loop/review-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 | $0.00034 | $0.01148 |
| Opus 5 | $0.00017 | $0.00574 |
| Sonnet 5 | $0.00007 | $0.00230 |
| Haiku 4.5 | $0.00003 | $0.00115 |
Grade A, and why
review-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 8d 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 — 84 lines — stays where its author put it; the contents beside it link to each section on GitHub.
the-loop: review-pr $ARGUMENTS
Review the change this thread is about, and change nothing yourself. The process is the
review loop (pdlc-review-loop), the fifth shipped graph: brief, review, follow-ups.
Ask the-loop check <id> / the-loop graph show where the item stands rather than
re-deriving it. If the reviewed repository carries a .the-loop/harness-config.yaml,
load it — it names the project's test, lint and type-check commands, which is what the
brief's Validations: usually mean; if it carries none, use the project's own visible
tooling and say which commands you ran.
Before acting, read the the-loop skill — the rules that are not about phases
still apply in full: the paper trail, the self-authored marker on every comment you post
(reference/collaboration.md § loop prevention), and context management. Only the
authorship is absent — see rule 1.
What this loop is
An authorized user armed this thread with the-loop review. That is their explicit,
recorded request for a review — frozen in graph-state.json's loop field, with the
arming comment standing on the thread. The loop will not review until the same class of
user states a brief (their questions, angles and validations); once one is frozen,
every round answers to it.
review-brief → review ⇄ follow-up → complete
review-brief— the-loop posts the fill-in template; an authorized reviewer answers it (the arming comment itself may already contain it). The parsed brief is frozen with their name on it.review— one round: answer every question, examine every angle, run every validation. Post the round as one self-marked comment on the thread.follow-up— the reviewer replies. Anything that is not "done" is another round, routed straight back toreviewwith the new reply as added scope. Silence leaves the gate open.complete— the reviewer said it is finished. A closed/merged thread ends it too.
The rules of a review
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.
- 8d ago First seen · 84 lines · 34 tokens per session scan A ffacf6a7e246
review-pr is a command published in the GitHub repository MadaraUchiha-314/the-loop (4 stars, last pushed 4d ago), licensed MIT. It adds 34 tokens to every session and 1,148 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 2026-08-31.
Other commands, from other repositories
reflect
Have an AI agent critique and improve its own work before delivering it, catching errors and gaps a first pass misses. Run like a /reflect command over a draft, plan, or answer.
critique
Get honest, useful critical feedback on any piece of work, with strengths, weaknesses, and how to fix them. A quick /critique command.
pr-description
Write a pull request description that helps reviewers: what changed, why, and how to verify, from a diff.
refactor-code
Suggest refactorings that improve readability and structure without changing behavior, ranked by payoff.
review-my-code
Review code for bugs, security issues, and clarity, with severity-ranked findings and concrete fixes.
review-security
Review code or a design for security weaknesses, ranked by real exploitability rather than by checklist coverage.