Automated reviewers usually drown a PR in nitpicks until people stop reading them. This one is built to stay quiet unless it has something real.
sonnet4 daysGitHubOpenAIESLintGitHub Actions
Claude
93ROI
88Scale
$3.9k93Saved
ROI for
README.md
Why this subagent
Automated reviewers usually drown a PR in nitpicks until people stop reading them. This one is built to stay quiet unless it has something real.
It reads the diff with surrounding context, flags only correctness, security, and maintainability issues, and leaves style to the linter. Comments land on the right line with a concrete fix, grouped so they are easy to act on. When the change is clean it says nothing, which is how reviewers earn trust.
How it runs
Used at step 01 to kick off the pipeline.
Write
Used at step 01 to kick off the pipeline.
WebFetch
Used at step 01 to kick off the pipeline.
WebSearch
Used at step 01 to kick off the pipeline.
Flag only correctness, security, and clear maintainability issues, skipping style a linter already covers.
pending
Group related comments and write each as a specific, actionable suggestion on the right line.
pending
Post the review and a short summary, holding back when there is nothing worth saying.
pending
Sample output
json
// Sample output
// (generated when the pipeline finishes)
Given a pull request diff and context, return a JSON list of line-level review comments limited to correctness, security, and maintainability, each with a concrete suggestion.
Unlock the rest
The full agent definition, install snippet, and starter task are gated for community members.
Members get the full `.md` agent file, the npm / pnpm install one-liners, a starter prompt that we've tuned against real runs, and the open-source repo when this automation ships there. One email, magic link, done.