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 agentmods add skills/baz-scm/baz-plugin/reviewnpx skills add baz-scm/baz-plugin --skill reviewgit clone --depth 1 https://github.com/baz-scm/baz-pluginWrote 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/baz-scm/baz-plugin/review)<a href="https://agentmods.dev/skills/baz-scm/baz-plugin/review"><img src="https://agentmods.dev/badge/skills/baz-scm/baz-plugin/review.svg" alt="Measured on agentmods" 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 | $0.00109 | $0.03154 |
| Opus 5 | $0.00055 | $0.01577 |
| Sonnet 5 | $0.00022 | $0.00631 |
| Haiku 4.5 | $0.00011 | $0.00315 |
Grade A, and why
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 5d 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 — 149 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Review with Baz
Review a set of changes and report what is actually wrong with them. $ARGUMENTS carries the scope flags (all optional).
What makes this review different from reading the diff yourself: Baz's indexed search lets you check the change against code that isn't in this checkout. A signature change that looks safe here can break a caller in another repo. Finding that is the point of this command — see Step 3.
Reviewing is read-only. Do not edit files, amend commits, or push. The one exception is the --fix loop in Step 6, which runs only after the user approves specific findings.
Step 1: Resolve the scope
Parse $ARGUMENTS, then build the diff.
Resolve the base branch first, and confirm the ref actually resolves before you diff against it — --base <branch> overrides all of this. Walk the ladder in order, taking the first that resolves under git rev-parse --verify:
git symbolic-ref refs/remotes/origin/HEAD— the remote's declared default. Commonly absent in a fresh or CI clone; that is expected, fall through.origin/main, thenorigin/master.main, thenmaster(local branches, for a repo with no remote).- Nothing resolved — you are most likely in a shallow or
--single-branchclone, where the base branch simply isn't in the object store. Do not diff againstHEAD~1and hope. Rungit fetch --depth=100 origin <branch>to pull the base in, or ask the user which base to use. Say which one you did.
Never assume main exists because the repo looks like it should have one — rev-parse --verify it. Diffing against a ref that doesn't resolve produces a git error mid-review; diffing against the wrong ref silently reviews the wrong changes, which is worse.
| Argument | What to review | How to get it |
|---|---|---|
| (none) | Everything not yet on the base branch — commits on this branch plus uncommitted tracked edits | git diff $(git merge-base <base> HEAD) |
committed |
Only commits on this branch | git diff <base>...HEAD |
uncommitted |
Only staged + unstaged edits to tracked files | git diff HEAD |
--include-untracked |
Adds new files git isn't tracking yet | git ls-files --others --exclude-standard, then read each |
--pr <number|url> |
An open pull request | gh pr diff <n> (or glab mr diff <n>) |
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.
- 5d ago First seen · 149 lines · 109 tokens per session scan A a7f0740f18b0
review is a skill published in the GitHub repository baz-scm/baz-plugin (10 stars, last pushed 8d ago), licensed MIT. It adds 109 tokens to every session and 3,154 once invoked, about $0.0005 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
taiyi-ui-design
TaiyiForge 第 4 阶段 — UI/UX 契约，产出 UI-DESIGN.md。四端通用。.
flow-next-resolve-pr
Resolve PR review feedback. Fetches unresolved threads, triages, fixes, replies and resolves via GraphQL. Use when asked to address review comments.
flow-next-tracker-sync
Project a flow-next spec to a tracker issue (Linear, GitHub, GitLab, Jira) and reconcile two-way. Use when asked to sync to a tracker. NOT plan-sync.
taiyi-evolve
TaiyiForge 辅助 — 实现后架构与文档同步（architecture-sync）。OpenCode / Claude / Codex / Cursor 通用。.
taiyi-diagram-c4
TaiyiForge 辅助 — 从代码反推 C4 架构文档（Observed/Inferred 分层 · Mermaid 真源）。OpenCode / Claude / Codex / Cursor 通用。.
flow-next-prime
Assess codebase agent and production readiness. Classifies the project, verifies commands run, leads with a verdict and ranked next actions.