[PR #6] [MERGED] docs(cursor-review): add directory README with overview + install guide #18

New issue

Closed

opened 2026-10-26 22:35:26 +00:00 by bsakaguchi · 0 comments

bsakaguchi commented

2026-10-26 22:35:26 +00:00

Owner

📋 Pull Request Information

Original PR: https://github.com/Comfy-Org/github-workflows/pull/6
Author: @mattmillerai
Created: 6/3/2026
Status: ✅ Merged
Merged: 6/3/2026
Merged by: @mattmillerai

Base: main ← Head: docs/cursor-review-readme

📝 Commits (1)

c5479c2 docs(cursor-review): add directory README with overview + install guide

📊 Changes

1 file changed (+180 additions, -0 deletions)

View changed files

➕ .github/cursor-review/README.md (+180 -0)

📄 Description

Summary

Adds .github/cursor-review/README.md — a higher-level overview and install guide for the multi-model Cursor Review panel, aimed at people who land in this directory and want to understand or adopt the review without reverse-engineering the workflow YAML.

What's covered

Overview — why a panel of frontier models from four labs + a judge beats trusting one model (broader coverage, less noise, surfaces what labs agree on).
How it works — an ASCII flow diagram: label → gate → 8-cell panel (4 labs × adversarial/edge-case) → judge consolidation → one PR review with severity badges.
The panel — the lab/model table and the two review types, linked to their prompts.
File map — what each prompt/script in this directory does.
Adopt it in your repo — the thin-caller workflow, the secrets/vars table (CURSOR_API_KEY, optional SLACK_BOT_TOKEN, CURSOR_REVIEW_DM_EMAIL_MAP), how to trigger, and the optional review-on-assignment companion.
Config knobs, escape hatches (skip-cursor-review, push-to-re-review, dismiss-to-rerun), and customizing (it's all source-of-truth here, so tuning is a normal PR).

Notes

Docs-only; no workflow behavior changes.
Cross-links the top-level README for pinning/versioning rather than duplicating it.
All claims verified against the workflow + scripts (model matrix, severity values critical/high/medium/low/nit, the DM email-map var, the panel summary line).

🤖 Generated with Claude Code

_{🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.}

## 📋 Pull Request Information **Original PR:** https://github.com/Comfy-Org/github-workflows/pull/6 **Author:** [@mattmillerai](https://github.com/mattmillerai) **Created:** 6/3/2026 **Status:** ✅ Merged **Merged:** 6/3/2026 **Merged by:** [@mattmillerai](https://github.com/mattmillerai) **Base:** `main` ← **Head:** `docs/cursor-review-readme` --- ### 📝 Commits (1) - [`c5479c2`](https://github.com/Comfy-Org/github-workflows/commit/c5479c2f3004166e2ba87b4064b72f9abce4a4ac) docs(cursor-review): add directory README with overview + install guide ### 📊 Changes **1 file changed** (+180 additions, -0 deletions) <details> <summary>View changed files</summary> ➕ `.github/cursor-review/README.md` (+180 -0) </details> ### 📄 Description ## Summary Adds `.github/cursor-review/README.md` — a higher-level overview and install guide for the multi-model Cursor Review panel, aimed at people who land in this directory and want to understand or adopt the review without reverse-engineering the workflow YAML. ## What's covered - **Overview** — why a panel of frontier models from four labs + a judge beats trusting one model (broader coverage, less noise, surfaces what labs agree on). - **How it works** — an ASCII flow diagram: label → gate → 8-cell panel (4 labs × adversarial/edge-case) → judge consolidation → one PR review with severity badges. - **The panel** — the lab/model table and the two review types, linked to their prompts. - **File map** — what each prompt/script in this directory does. - **Adopt it in your repo** — the thin-caller workflow, the secrets/vars table (`CURSOR_API_KEY`, optional `SLACK_BOT_TOKEN`, `CURSOR_REVIEW_DM_EMAIL_MAP`), how to trigger, and the optional review-on-assignment companion. - **Config knobs**, **escape hatches** (`skip-cursor-review`, push-to-re-review, dismiss-to-rerun), and **customizing** (it's all source-of-truth here, so tuning is a normal PR). ## Notes - Docs-only; no workflow behavior changes. - Cross-links the top-level README for pinning/versioning rather than duplicating it. - All claims verified against the workflow + scripts (model matrix, severity values `critical/high/medium/low/nit`, the DM email-map var, the panel summary line). 🤖 Generated with [Claude Code](https://claude.com/claude-code) --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>

bsakaguchi

2026-10-26 22:35:26 +00:00