sw-review

sw-review is a skill for Claude Code from anton-abyzov/specweave. It costs 42 tokens per session (941 once invoked), scanned A, original, MIT.

A fresh-context code review skill that looks for concrete implementation failures before a task ships. Each reported issue includes a file path and line number and is checked again before being recorded.

In plain words
What is it for?
Use it to review an increment, report reproducible problems with severity levels, rerun relevant checks, and save the findings in a review report.
Why use it?
It provides an independent check instead of relying on the person who wrote the code to approve it. This helps separate real bugs from style opinions or unverified concerns.

Skill for Claude Code

Written for Claude Code: argument-hint in frontmatter. Also seen: mentions subagents.

Good fit Use it to review an increment, report reproducible problems with severity levels, rerun relevant checks, and save the findings in a review report.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/anton-abyzov/specweave/sw-review
Install

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.

Any agent
npx skills add anton-abyzov/specweave --skill sw-review
Clone the repo
git clone --depth 1 https://github.com/anton-abyzov/specweave

Made for: Claude Code.

Wrote 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.

agentmods badge for sw-review

README.md
[![agentmods](https://agentmods.dev/badge/skills/anton-abyzov/specweave/sw-review/github.svg)](https://agentmods.dev/skills/anton-abyzov/specweave/sw-review)
Your own site
<a href="https://agentmods.dev/skills/anton-abyzov/specweave/sw-review"><img src="https://agentmods.dev/badge/skills/anton-abyzov/specweave/sw-review/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.

agentmods 80×15 button for sw-review

Your own site · 80×15
<a href="https://agentmods.dev/skills/anton-abyzov/specweave/sw-review"><img src="https://agentmods.dev/badge/skills/anton-abyzov/specweave/sw-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 42 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 941 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00042 $0.00941
Opus 5 $0.00021 $0.00470
Sonnet 5 $0.00008 $0.00188
Haiku 4.5 $0.00004 $0.00094

Measured 8d ago against content hash 8fb2951a8b5a, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

sw-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 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.

skills/sw-review/SKILL.md · 66 lines

How it starts

The opening of the file, as written. The whole thing — 66 lines — stays where its author put it; the contents beside it link to each section on GitHub.

sw-review — adversarial review with citations

The only review gate that changes code: a fresh reader, hunting for real failures, citing path:line, and re-checking every claim before writing it down. Writes .specweave/increments/<id>/reports/review.md.

Non-negotiables

  • Fresh context. The session that wrote the code never approves it. Start a new conversation (or a sub-agent, in tools that have them). If you wrote this code in this session, say so and stop — the review is not credible.
  • Every finding cites path:line and a concrete failure: inputs or state → wrong output, crash, hang, corruption or leak. "Consider refactoring" is not a finding.
  • Re-verify before reporting. Re-open the cited lines and, where a test can show it, run one. Drop anything that does not survive. A wrong finding costs more than a missed one.
  • Severity: critical / high / medium / low. Only critical and high should block a close.
  • Nits: at most 5, collected in one line at the end. Style opinions are not review output.

No CLI? Nothing changes: git diff and the project's test runner are the whole toolchain, and you write reports/review.md by hand. The manual path and the CLI path produce the same report.

Steps

  1. Scope — read spec.md (ACs + Approach) and the Files: fields in tasks.md. Review that diff, not the repository:

    git diff $(git merge-base HEAD origin/HEAD)...HEAD -- <the tasks' Files>
    

    Falling back when there is no merge base: git diff HEAD~1...HEAD.

  2. Pass — in this order: correctness (wrong logic, unhandled errors, races, off-by-one, injection, secret leakage, data loss) → ACs claimed but not implemented → tests that assert nothing (no expectation, mocked subject, snapshot of the bug) → cross-platform breakage (bash-only code, \ paths, BOM, >> in PowerShell). --full: run three passes — correctness, security, spec-compliance — and merge, dropping duplicates.

  3. Verify — for each candidate, re-read the lines and state the failure scenario in concrete terms. If you cannot, delete it.

  4. Report — write reports/review.md in the shape below. A machine-readable sibling reports/review.json is optional: {"ok":false,"findings":[{"severity":"critical","file":"src/a.ts","line":142,"summary":"…"}]}.

  5. Hand back — critical/high go back to sw-do as tasks (or get fixed and re-tested), then specweave verify <id> again before specweave complete <id>.

Read the full file on GitHub · 66 lines

Changes

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.

  1. 8d ago First seen · 66 lines · 42 tokens per session scan A 8fb2951a8b5a

Subscribe to this mod's changes

sw-review is a skill published in the GitHub repository anton-abyzov/specweave (160 stars, last pushed today), licensed MIT. It adds 42 tokens to every session and 941 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 2026-09-03.