A code-review workflow that compares recent changes with decisions recorded in Kauri, a project decision database. It identifies relevant decisions and reports whether the code follows them, conflicts with them, or makes them outdated.

Use it after implementation, before committing, or during a pull-request review. It can inspect file paths, commit ranges, and current uncommitted changes.

It catches changes that contradict agreed constraints before they are committed or reviewed. It also highlights decisions that may need replacement when the design has intentionally changed.

review — Skill for Claude Code by mhashim6/kauri

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 mhashim6/kauri --skill review

Clone the repo

git clone --depth 1 https://github.com/mhashim6/kauri

Made for: Claude Code.

Or install kauri, the plugin that ships this one along with the rest of its 5 skills, 2 hooks, 1 MCP server.

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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/mhashim6/kauri/review/github.svg)](https://agentmods.dev/skills/mhashim6/kauri/review)

Your own site

<a href="https://agentmods.dev/skills/mhashim6/kauri/review"><img src="https://agentmods.dev/badge/skills/mhashim6/kauri/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.

Your own site · 80×15

<a href="https://agentmods.dev/skills/mhashim6/kauri/review"><img src="https://agentmods.dev/badge/skills/mhashim6/kauri/review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 39 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 545 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.

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.

Model	Per session	Once invoked
Fable 5.1	$0.00039	$0.00545
Opus 5.5	$0.00016	$0.00218
Sonnet 5.5	$0.00008	$0.00109
Haiku 4.5	$0.00004	$0.00055

Measured 1mo ago against content hash f99f7069fd9e, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

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 1mo 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.

plugins/claude-code/skills/review/SKILL.md · 53 lines

What it actually says

Review Changes Against Decisions

After implementing changes, verify they are consistent with recorded project decisions. This catches violations before they're committed.

Steps

Identify what changed: If $ARGUMENTS is a file path, read it. If it's a commit range, run git diff $ARGUMENTS. If empty, run git diff to see unstaged changes, or git diff --cached for staged changes.
Find related decisions: For each changed file, query Kauri:
- kauri_query with --file <path> for directly associated decisions.
- kauri_query with --text using keywords from the changed code (function names, module names, concepts).
Read the full decisions: Call kauri_show <id> for every match. Don't skip any — a decision might look irrelevant from the title but contain a constraint in the body.
Assess consistency: For each relevant decision, determine:
- Consistent: The changes respect the decision. Note this.
- Potential conflict: The changes might violate the decision. Explain specifically what conflicts and why.
- Decision should be updated: The changes are intentional improvements that make the old decision obsolete.
Report to the user:
- List each relevant decision and your assessment.
- For conflicts: recommend either adjusting the code or superseding the decision.
- For decisions that should be updated: offer to record a superseding decision via /kauri:propose.
- For decisions now validated by the changes: offer to run kauri_validate still_valid to refresh their timestamp.

When to use this

Before git commit on significant changes
When reviewing a PR or diff
When the user asks "does this look right?" or "anything I'm missing?"
After refactoring that touches many files
When the post-commit staleness hook reports stale decisions

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.

1mo ago First seen · 53 lines · 39 tokens per session scan A f99f7069fd9e

Subscribe to this mod's changes

review is a skill published in the GitHub repository mhashim6/kauri (5 stars, last pushed 6mo ago), licensed MIT. It adds 39 tokens to every session and 545 once invoked, about $0.0002 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 2026-08-31.

Code review adr agent agent-memory ai-agents calude decision

review

Getting it into your agent

Wrote this? Show the measurements

What it costs to keep this loaded

Grade A, and why

Nothing flagged

What it actually says

Review Changes Against Decisions

Steps

When to use this

What this file has done since we first saw it

Other skills, from other repositories

code-review

github-code-review

simplify-code

requesting-code-review

ADR Decision Quality

pragma-code-review