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 commands/semihkayan/codeweave-mcp/reviewgit clone --depth 1 https://github.com/semihkayan/codeweave-mcpWrote 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/commands/semihkayan/codeweave-mcp/review)<a href="https://agentmods.dev/commands/semihkayan/codeweave-mcp/review"><img src="https://agentmods.dev/badge/commands/semihkayan/codeweave-mcp/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 | $0.00006 | $0.00267 |
| Opus 5 | $0.00003 | $0.00133 |
| Sonnet 5 | $0.00001 | $0.00053 |
| Haiku 4.5 | $0.00001 | $0.00027 |
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.
What it actually says
You should start a review process for the implementation of a plan.
1. Spawn The Reviewer
Replace with the actual path to the plan file. Do not pass the plan content. Agent( subagent_type: "plan-implementation-reviewer", model: "opus", description: "Review The Implementation", prompt: "Review the implementation of this plan: " )
2. Understand The Findings
The reviewer will return a list of findings and its agentId. Analyze the report. The findings do not need to be correct. Scan for false positives.
3. Resume The Reviewer
Do not ask to the user. Do not fix yourself. Resume the reviewer via SendMessage tool with agentId. Give it the ones you want it to address and additional context if needed.
Exception: If the total fix work is less than 15 lines, fix yourself.
4. Review The Fixes
Review the fixes proposed by the reviewer. If plan verification steps exist, repeat them.
5. Report
Report the final state to the user. Concise. Max 30 words per topic. Which findings are addressed, follow-ups, considerations, etc.
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 · 28 lines · 6 tokens per session scan A cc59f4fca6a2
review is a command published in the GitHub repository semihkayan/codeweave-mcp (4 stars, last pushed 3mo ago), licensed Apache-2.0. It adds 6 tokens to every session and 267 once invoked, about $0.0000 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 commands, from other repositories
feature
Add a feature the way this repo requires — spec mapped to STANDARD.md, failing tests first, harness tiers, bench delta, independent review, and a machine-produced Definition-of-Done checklist; opens the PR only when nothing is unmet.
release
Prepare a release the way RUNBOOK section 1 requires — confirm main is green, derive the next version and show the derivation, reconcile the changelog against merged PRs, recompute every published figure and refuse on a disagreement, draft the notes, open the labelled release PR, then STOP. Merging and dispatching are…
investiate-debug-task
Please investigate this task, debug it (using scripts in /tmp if needed) in order to find the root cause of the issue. Find the and suggest the best strategies for fixing the issue.
fix-issue
Fix a GitHub issue — reproduce it as a failing test before touching source (stop if it cannot be reproduced), check ARCHAEOLOGY for the guard that should have caught it, fix minimally, run the tiers, independent review, changelog, PR referencing the issue, machine-produced checklist.
review
Grade a PR, a ref, or the working tree against the Definition of Done in an isolated reviewer subagent. --merge-check adds the contributor-PR gate — trial merge onto origin/main in a scratch worktree, fast tier there, the diff's DELETIONS first, license/cla count on the head SHA. Verdict to chat and…
feedback
Security Design Review — PRD/기획서 기반 보안 의견서·검토 의견서 생성.