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/LeanAndMean/scramjetWrote 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/leanandmean/scramjet/mach12-pr-review-assessment)<a href="https://agentmods.dev/commands/leanandmean/scramjet/mach12-pr-review-assessment"><img src="https://agentmods.dev/badge/commands/leanandmean/scramjet/mach12-pr-review-assessment/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/leanandmean/scramjet/mach12-pr-review-assessment"><img src="https://agentmods.dev/badge/commands/leanandmean/scramjet/mach12-pr-review-assessment.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 | $0.00013 | $0.06258 |
| Opus 5 | $0.00006 | $0.03129 |
| Sonnet 5 | $0.00003 | $0.01252 |
| Haiku 4.5 | $0.00001 | $0.00626 |
Grade A, and why
mach12:pr-review-assessment 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 6d 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 — 305 lines — stays where its author put it; the contents beside it link to each section on GitHub.
PR Review Assessment
Goals
- Independently classify every review finding by whether the problem is real and whether a sound correction would improve the combined system.
- Publish one complete, verified assessment with supported dispositions, rejected regressions, and bounded follow-up scope.
- After verified assessment publication, carry out only user-authorized deferred-item publications or metadata changes and durably record their actual outcomes.
- Route only authenticated findings that remain worth fixing, or proceed to pre-merge when no required correction remains.
Step 1: Parse input
The user's input typically contains:
- A PR number (required)
- A
--review-comment <id>flag with a numeric comment ID (optional) - Additional context (optional)
Extract the PR number and the --review-comment ID if present. If the input is ambiguous, ask the user to clarify.
Step 2: Gather PR and review context
Locate the review comment
If --review-comment was provided: Fetch the specific comment by ID, then fetch the PR title, body, and all comments for context.
gh api repos/:owner/:repo/issues/comments/<review-comment-id>
Extract the body, numeric ID, author login, and URL from the JSON response. Then delegate to /mach12:gh-pr-read <pr-number> (no marker) to fetch the complete verified PR context. Require the explicit ID to match exactly one comment in that target PR's stream, require the comment to contain <!-- mach12-review --> or the recognized legacy review structure, and require its author to match the authenticated gh api user --jq .login identity. Stop before assessment when any check fails.
If --review-comment was NOT provided: Delegate to:
/mach12:gh-pr-read <pr-number> --marker mach12-review
The subroutine returns the PR title, body, full comments array, and the matched review comment body and numeric ID (using the most recent marker match). If no comment contains the marker, the subroutine reports that and the caller falls back to the last comment with the structured review format (Critical/Important/Suggestions sections and model attribution).
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.
- 6d ago Changed · +12 lines fe53c5ee584e
- 11d ago First seen · 293 lines · 13 tokens per session scan A 396c89d04c25
mach12:pr-review-assessment is a command published in the GitHub repository LeanAndMean/scramjet (7 stars, last pushed today), licensed Apache-2.0. It adds 13 tokens to every session and 6,258 once invoked, about $0.0001 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
computer
Drive a computer's screen, mouse, and keyboard.
find-issues
Find GitHub issues that a PR might fix.
afr
Review feature - code review with fixes by a different agent (shortcut for feature-code-review).
afrv
Revise the current feature worktree after code review — decide accept/revert/modify (shortcut for feature-code-revise).
afe
Evaluate feature - code review or comparison (shortcut for feature-eval).
find-duplicate-prs
Find open PRs that may duplicate a given PR.