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 LuisUrrutia/skills --skill review-code-changesgit clone --depth 1 https://github.com/LuisUrrutia/skillsWrote 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/luisurrutia/skills/review-code-changes)<a href="https://agentmods.dev/skills/luisurrutia/skills/review-code-changes"><img src="https://agentmods.dev/badge/skills/luisurrutia/skills/review-code-changes/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/luisurrutia/skills/review-code-changes"><img src="https://agentmods.dev/badge/skills/luisurrutia/skills/review-code-changes.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 | $0.00028 | $0.01841 |
| Opus 5.5 | $0.00011 | $0.00736 |
| Sonnet 5.5 | $0.00006 | $0.00368 |
| Haiku 4.5 | $0.00003 | $0.00184 |
Grade A, and why
review-code-changes 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 — 153 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Review code changes
Review one declared change and deliver findings supported by its actual code, requirements, and contracts. Apply the project's language and runtime semantics. This phase preserves product files, tests, the index, HEAD, and remote state. Write reports and permitted probes in the caller's output or scratch location. An enclosing implementation task can act on the findings after this review ends.
Read references/protocol.md before reviewing. It owns
the review criteria, evidence rules, severity, and canonical report grammar.
A caller that runs this audit in an engine unable to load skills may freeze that
protocol and scripts/report.py unchanged into its run packet, supplying its own
single-auditor role in place of this entrypoint's orchestration.
Establish a reproducible scope
Honor the supplied comparison. Resolve refs to commit IDs and record the commands and path inventory before reviewing:
- For a branch or PR, establish its actual target base from the caller or PR/repo metadata and use its merge-base with the head. An upstream tracking branch may be the PR's own head. Exclude earlier layers of a stack and unrelated base work.
- For a commit, inspect that commit's patch. Resolve the intended parent for a merge commit when different choices would change the review.
- For local changes, inspect staged and unstaged diffs and relevant untracked files. Record both intermediate and final states when staging changes behavior.
- For a supplied diff or file list, keep that exact boundary; surrounding code supplies evidence, not permission for a whole-repository audit.
For an unspecified scope, use current worktree changes when present; otherwise resolve the current branch's target base from accessible context. Ask only when the unresolved choice materially changes what is reviewed. An empty scope is a valid result: report it instead of inventing work.
Save the unique changed-path inventory as a JSON array for report validation. Record deleted paths and the new path of a rename, identifying the old path in Coverage. A submodule bump is a pin change: record its old and new commits and URL, and review the superproject's integration with the new pin; the submodule's own changes are outside scope unless the caller adds them. Capture Git state so concurrent edits or accidental mutations are detectable. If the input changes during review, refresh the affected analysis or report that the result covers the earlier snapshot.
What ships with it
12 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.
- assets/report.css 800 B
- origin.toml 76 KB
- references/delegation.md 5.3 KB
- references/profile-schema.md 3.2 KB
- references/profiles.md 2.8 KB
- references/protocol.md 27 KB
- references/specialists.md 2.5 KB
- scripts/report.py 13 KB runs code
- scripts/resolve_profile.py 3.9 KB runs code
- tests/test_profiles.py 3.6 KB runs code
- tests/test_report.py 6.5 KB runs code
- THIRD-PARTY-NOTICES.txt 8.5 KB
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 · +30 lines f4cba3254f70
- 17d ago Changed · -3 lines 3001d4e2d105
- 19d ago Changed · +2 lines 31467f02c5b3
- 25d ago Changed · +3 lines c93a230256fc
- 1mo ago First seen · 121 lines · 28 tokens per session scan A 731b7c5cc7b9
review-code-changes is a skill published in the GitHub repository LuisUrrutia/skills (2 stars, last pushed today), licensed MIT. It adds 28 tokens to every session and 1,841 once invoked, about $0.0001 per session on Opus 5.5. 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 2026-10-06.
Other skills, from other repositories
code-review-and-quality
Use when reviewing Android code (own or others'). Five-axis review framework: Correctness, Readability, Architecture, Security, Performance. Categorized findings with Kotlin/Compose-specific checks.
code-simplification
Use when code is harder to understand than it needs to be. Guides incremental simplification that preserves behavior exactly, following project conventions and Kotlin idioms.
doubt-driven-development
Use when a decision is high-stakes and hard to reverse — architecture choices, data migrations, dependency adoption, public API contracts. Adversarial self-review that attacks the chosen approach before the code does, instead of defending the first idea that worked.
fw-review
Full Freshworks marketplace app review — iparams, frontend, serverless, FDK, security, and structured text report output — in one skill.
battle-tested-engineer
Engineering judgment for code write/refactor/test and frontend UI. Trigger on code review, legacy cleanup, tests, UI with data, bloat/over-engineer complaints — even with no explicit mention, before any diff/rewrite/test suite. Output ultra-terse caveman style; code, commits, PR desc, security warnings stay normal…
agent-watchdog
Use when asked to watch, babysit, audit, review, compare, or fix another agent's work from a Codex session ID, Claude Code session/transcript, chat/thread link, PR, branch, log, or pasted run summary. Monitor until the other agent is done or blocked, reconstruct what the user asked, independently investigate the same…