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 sleeyax/promptfiles --skill review-mrgit clone --depth 1 https://github.com/sleeyax/promptfilesWrote 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/sleeyax/promptfiles/review-mr)<a href="https://agentmods.dev/skills/sleeyax/promptfiles/review-mr"><img src="https://agentmods.dev/badge/skills/sleeyax/promptfiles/review-mr/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/sleeyax/promptfiles/review-mr"><img src="https://agentmods.dev/badge/skills/sleeyax/promptfiles/review-mr.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 | $0.00043 | $0.02748 |
| Opus 5 | $0.00022 | $0.01374 |
| Sonnet 5 | $0.00009 | $0.00550 |
| Haiku 4.5 | $0.00004 | $0.00275 |
Grade A, and why
review-mr 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 10d 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 — 171 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Review a Merge Request
MR: $ARGUMENTS
Review a GitLab MR like a senior engineer on the team and post findings as draft notes so the user publishes/discards manually.
This skill fetches the MR's context and places the comments. The reviewing itself is delegated to the review-changes skill in report-only mode, which picks the reviewers (two agents by default), runs them, and returns one merged, deduplicated set of findings.
Hard rules
- Never
publishdraft notes — leave them for the user. - Every gate — dirty-tree, no-checkout, diff-size, pre-submission — is a real stop: ask, then wait for the answer. Use the
AskUserQuestiontool when it's available in the session; where it isn't (e.g. Codex), ask in plain text with the same numbered options and stop until the user replies. Never assume an answer. - Never push, commit, or modify the cwd repo's working tree without an explicit confirm.
- Every finding cites a real
+(added) line in the actual diff. Removed-line comments are out of scope. - Posting is gated by exactly one explicit confirm covering the whole batch.
- Don't review the diff here. Delegate to
review-changesand work from the findings it returns; the only exception is diff-only mode, where that skill can't run — and the user has to pick it at the step 5 gate. - One defect is one draft note.
review-changesmerges and deduplicates before returning — never re-split a merged finding, and never post the same defect twice because both agents found it. - The agent writes exactly one
findings.jsonper run (matchingfindings.schema.json) and invokespost-draft-note.shexactly once. No per-finding shell calls. Rollback on partial failure is the helper's job. - GitLab MCP server first if available; fall back to
glab. If neither, stop. - Never dump the full diff to a file or to chat, or hand it to a subagent as one blob. When the cwd repo matches the MR, read the diff incrementally from
git(per-file, on demand) — not fromglab api .../diffs. The APIdiffsendpoint is only used when there is no local checkout, and even then read it page-by-page, one file at a time, never as one bulk write. To measure size up front, use/merge_requests/<iid>/changes(metadata + per-file stats), not/diffs. This rule still holds when the user picks Review whole diff anyway — that answer authorizes the scope, not bulk-dumping.
What ships with it
4 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.
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.
- 10d ago First seen · 171 lines · 43 tokens per session scan A 029fa4e8a23a
review-mr is a skill published in the GitHub repository sleeyax/promptfiles (2 stars, last pushed 11d ago), licensed MIT. It adds 43 tokens to every session and 2,748 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 skills, from other repositories
autoreview
Pre-commit/ship code review: Codex default; optional Claude or Pi.
rework-rate
Measure and interpret PR rework rate — the emerging 5th DORA metric.
omh-code-review
This is a Hermes-native code-review workflow skill.
revdiff-plan
Review the last Codex assistant message (plan, analysis, or proposal) with inline annotations in a TUI overlay. Extracts the most recent response from Codex rollout files and opens it in revdiff for review and annotation. Activates on "revdiff-plan", "review plan with revdiff", "annotate plan", "review last response"…
code-reviewer
Code review specialist focused on patterns, bugs, security, and performance.
full-repo-review
Comprehensive four-wave review of all repo source files, producing a prioritized issue backlog.