15 Commits

Author SHA1 Message Date
Chris Lloyd
a93966285e Unify issue lifecycle labeling and sweep into a single system (#25352)
* Unify issue lifecycle labeling and sweep into a single system

Consolidate issue triage, stale detection, and lifecycle enforcement into
two components: a Claude-powered triage workflow and a unified sweep script.

Triage workflow changes:
- Add issue_comment trigger so Claude can re-evaluate lifecycle labels when
  someone responds to a needs-repro/needs-info issue
- Add concurrency group per issue with cancel-in-progress to avoid pile-up
- Filter out bot comments to prevent sweep/dedupe triggering re-triage
- Hardcode allowed label whitelist to prevent label sprawl (was discovering
  labels via gh label list, leading to junk variants like 'needs repro' vs
  'needs-repro')
- Replace MCP GitHub server with gh CLI — simpler, no Docker dependency,
  chaining is caught by the action so permissions are equivalent
- Add lifecycle labels (needs-repro, needs-info) for bugs missing info
- Add invalid label for off-topic issues (Claude API, billing, etc.)
- Add anti-patterns to prevent false positives (don't require specific
  format, model behavior issues don't need traditional repro, etc.)

Sweep script changes:
- Absorb stale issue detection (was separate stale-issue-manager workflow)
- Mark issues as stale after 14 days of inactivity
- Skip assigned issues (team is working on it internally)
- Skip enhancements with 10+ thumbs up (community wants it)
- Add invalid label with 3-day timeout
- Add autoclose label support to drain 200+ legacy issues
- Drop needs-votes (stale handles inactive enhancements)
- Unify close messages into a single template with per-label reasons
- Run 2x daily instead of once

Delete stale-issue-manager.yml — its logic is now in sweep.ts.

## Test plan

Dry-run sweep locally:
GITHUB_TOKEN=$(gh auth token) GITHUB_REPOSITORY_OWNER=anthropics   GITHUB_REPOSITORY_NAME=claude-code bun run scripts/sweep.ts --dry-run

Triage workflow will be tested by opening a test issue after merge.

* Update .github/workflows/claude-issue-triage.yml

Co-authored-by: Ashwin Bhat <ashwin@anthropic.com>

---------

Co-authored-by: Ashwin Bhat <ashwin@anthropic.com>
2026-02-12 11:45:31 -08:00
Chris Lloyd
a17040212c Add daily sweep to enforce issue lifecycle label timeouts
Introduce a simple, mechanical daily sweep that closes issues with
lifecycle labels past their timeout:

- needs-repro: 7 days
- needs-info: 7 days
- needs-votes: 30 days
- stale: 30 days

The sweep checks when the label was last applied via the events API,
and closes the issue if the timeout has elapsed. No AI, no comment
checking — if the label is still there past its timeout, close it.
Removing a label (by a triager, slash command, or future AI retriage)
is what prevents closure.

Each close message directs the reporter to open a new issue rather
than engaging with the closed one.

The script supports --dry-run for local testing:
  GITHUB_TOKEN=$(gh auth token) \
  GITHUB_REPOSITORY_OWNER=anthropics \
  GITHUB_REPOSITORY_NAME=claude-code \
  bun run scripts/sweep.ts --dry-run

## Test plan

Ran --dry-run against anthropics/claude-code. Correctly identified 3
issues past their timeouts (1 needs-repro at 12d, 2 needs-info at
14d and 26d). No false positives.
2026-02-11 22:34:23 -08:00
David Dworken
a3df424857 fix: validate that issues exist before commenting on duplicates
Add validation to comment-on-duplicates.sh that verifies the base issue
and all potential duplicate issues actually exist in the repo before
posting a comment.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>
2026-01-07 07:57:56 -08:00
Boris Cherny
0b86fdb0e0 fix(security): Remove overly broad gh api permission from dedupe command (#16549)
* fix(security): Remove overly broad gh api permission from dedupe command

Remove `Bash(gh api:*)` from dedupe.md allowed-tools to prevent potential
secret exfiltration via prompt injection. The dedupe workflow only needs
gh issue view/list/comment and gh search commands - it doesn't require
raw API access.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>

* feat: Add comment-on-duplicates script for safer duplicate handling

Replace `gh issue comment:*` permission with a constrained script that:
- Only accepts validated issue numbers
- Enforces max 3 duplicates
- Uses a fixed comment format
- Prevents arbitrary comment content injection

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-07 08:46:31 +05:30
Adam Wolff
09a020096a Restore dedupe workflows
This reverts commit d81efef324.
2025-09-24 18:01:05 -07:00
inigo
d81efef324 cleanup
🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>
2025-09-19 10:39:06 -07:00
Boris Cherny
20ba9a34a5 feat: update dedupe backfill script to filter by issue number
Replace date-based filtering with issue number filtering to only process issues older than #4050. This provides more precise control over which issues are processed for duplicate detection backfill.

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>
2025-08-18 09:54:40 -07:00
Boris Cherny
370a97d939 fix: improve duplicate issue number extraction in auto-close script
The extractDuplicateIssueNumber function now handles both #123 format
and full GitHub issue URLs like https://github.com/owner/repo/issues/123.
This fixes the "could not extract duplicate issue number from comment"
errors that were occurring when the script encountered URL-formatted
issue references in duplicate detection comments.

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>
2025-08-14 16:37:46 -07:00
Boris Cherny
4a04589002 Use proper 'duplicate' state_reason for issue closures
- Update auto-close script to use state_reason: 'duplicate' instead of 'not_planned'
- Simplify workflow detection logic to only check for duplicate state_reason
- Remove fallback logic for backward compatibility - use modern GitHub API

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>
2025-08-11 15:44:08 -07:00
Boris Cherny
04cace9ec0 Consolidate GitHub issue closure events to prevent duplicates
- Remove duplicate Statsig logging from auto-close-duplicates.ts
- GitHub workflow now handles all issue closures uniformly
- Add 'duplicate' label to ensure proper detection by workflow
- Prevents double-logging when script closes issues

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>
2025-08-11 15:33:31 -07:00
Boris Cherny
5af0b38a92 Add Statsig event logging to GitHub issue workflows
- Log events when issues are closed as duplicates in auto-close script
- Log events when duplicate comments are added via dedupe workflow
- Log events when new issues are created
- Follow existing pattern from code review reactions workflow

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>
2025-08-08 18:26:48 -07:00
Boris Cherny
d7aa61e6f1 Add script to backfill duplicate comments for old issues
Creates a script that identifies old issues without duplicate detection comments and triggers the existing claude-dedupe-issues workflow for each one. This helps ensure historical issues get proper duplicate detection coverage.

Features:
- Scans issues from configurable time period (default 30 days)
- Skips issues that already have duplicate detection comments
- Triggers existing workflow instead of duplicating logic
- Includes dry-run mode and rate limiting for safety

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>
2025-08-08 12:06:36 -07:00
Boris Cherny
1cc90e9b78 Update auto-close-duplicates script to actually close issues
- Remove dry run mode and implement actual issue closing
- Extract duplicate issue number from bot comments
- Close issues via GitHub API with proper state and comments
- Add error handling for API failures
- Use Claude Code comment format with reopening instructions

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>
2025-08-08 11:52:34 -07:00
Boris Cherny
eb48a37a48 Improve auto-close duplicates script pagination and filtering
- Add pagination to fetch more than 100 issues (up to 20 pages/2000 issues)
- Filter to only process issues created more than 3 days ago
- Add created_at field to GitHubIssue interface

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>
2025-08-08 11:21:27 -07:00
Boris Cherny
b7116c78d8 Extract auto-close duplicates script to TypeScript with Bun
- Move GitHub workflow script logic to scripts/auto-close-duplicates.ts
- Convert from inline JavaScript to standalone TypeScript module
- Update workflow to use Bun instead of Node.js for better performance
- Add proper TypeScript types and ES module syntax

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>
2025-08-08 11:11:41 -07:00