review
Get the change reviewed by the second model before reporting it done, post that review on the pull request, and act on every finding. Use after the gate passes and the pull request exists, before the DONE report.
Install / Use
npx skills add datahub-local/datahub-local-ai --skill reviewInstalls into whichever agent you are using.
SKILL.md
Installable skill definition
Quality Score
Category
Development & EngineeringSupported Platforms
Our assessment of review
review scores 54/100 on our quality scale, 4257th of 4,653 Development & Engineering skills we index.
Its SKILL.md is 2.4 KB long, split into 7 sections with 1 code example: moderately detailed.
It has no GitHub stars yet, so there is no community track record; judge it on its content.
Maintenance, license and trust
- We could not determine when the repository was last updated.
- No license is declared. By default that means all rights are reserved: you can read it, but reusing or redistributing it is not clearly permitted. Ask the author before building on it commercially.
- Its trust signals score 68/100, with 3 cautions from licensing, adoption, age or documentation. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.
Safety scan
No issues foundOur scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands.
Automated pattern scan on 2026-11-02. It catches known dangerous patterns, not every risk — read a skill before letting an agent act on it.
review compared with similar skills
All 4 of these similar skills score higher than review; compare them before choosing.
| Skill | Score | Stars | Updated | Format |
|---|---|---|---|---|
| review (this skill)by datahub-local | 54 | 0 | — | SKILL.md |
| taste-skillby Leonxlnx | 100 | 107.9k | 1d ago | CLAUDE.md |
| ai-job-searchby ai-job-search-dev | 100 | 68.8k | 1d ago | CLAUDE.md |
| claude-howtoby luongnv89 | 100 | 46.4k | today | CLAUDE.md |
| algorithmic-artby anthropics | 100 | 177.9k | 40d ago | SKILL.md |
Frequently asked questions
- How do I install review?
- Run
npx skills add datahub-local/datahub-local-ai --skill review. The install tabs above show the steps for each supported agent. - Which AI agents does review work with?
- It is written for Universal, as a SKILL.md file. Other agents that read the same format can often use it too.
- Is review safe to use?
- Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. It declares no license and scores 68/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
- Is review still maintained?
- We could not determine when the repository was last updated.
Skill content
View source on GitHubname: review description: Get the change reviewed by the second model before reporting it done, post that review on the pull request, and act on every finding. Use after the gate passes and the pull request exists, before the DONE report.
Review
The agent that wrote the change is the worst judge of it. A second agent, on a different model, reads the diff with fresh eyes, and its review is written where a person will find it: on the pull request.
Order
- The gate is green and the pull request exists.
- Run the reviewer (below).
- Post the review on the pull request.
- Act on every finding, then push.
- Report. At most two review rounds.
Never run this before the gate: a reviewer reading a failing diff wastes its turn on what the gate already said.
Running the reviewer
Invoke the reviewer subagent with the Task tool. It is the reviewer agent in the
image config: a different model (OPENCODE_REVIEW_MODEL), read-only by permission,
no question, no nested subagents. Give it:
- the issue number and the repository;
- the base the diff is against (
origin/main); - the gate commands you ran and their results;
- the files you changed, one line each.
Do not paste the diff or your reasoning: it reads them. Do not ask it to fix anything; it cannot edit.
Posting it
Write the review verbatim on the pull request, with a quoted heredoc so nothing in the body is interpreted:
cat <<'EOF' | gh pr comment <n> --repo datahub-local/<repo> --body-file -
## Review by glm-5.3-flash
<the reviewer's findings, verbatim>
<!-- taskboard:review -->
EOF
The marker matters: it is what tells the tick this comment is not the owner writing.
Acting on it
- BLOCKING: fix it. If you believe the finding is wrong, say why in one line in the report and in a reply on the pull request; do not stay silent and do not silently comply with something wrong.
- SHOULD: fix it, or state in the report why it is out of scope.
- NIT: fix it if it is free, otherwise leave it.
- Re-run the gate after the fixes and push to the same branch.
- Run the reviewer once more when a blocking finding was fixed. Two rounds is the ceiling; a third means the task is bigger than one branch, and that is a report.
In the report
One line per round: the verdict, the count of findings by severity, and what you
did with each. Unresolved findings are named with the reason. A review you did not
run is [UNVERIFIED], not omitted.
Related Skills
taste-skill
107.9kAgent skills that give coding agents taste: art direction, reference images and a browser check instead of generic AI pages
ai-job-search
68.8kThe job search that runs on your machine. AI job application framework built on Claude Code: evaluate postings, tailor CVs, write cover letters, prep interviews. Fork it and own it.
claude-howto
46.4kA visual, example-driven guide to Claude Code — from basic concepts to advanced agents, with copy-paste templates that bring immediate value.
algorithmic-art
177.9kCreating algorithmic art using p5.js with seeded randomness and interactive parameter exploration. Use this when users request creating art using code, generative art, algorithmic art, flow fields, or particle systems.
Trust signals
From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.
