github.com/jeffvli/feishin (link)

commit signals (14)

• AiGenerationPhrase
  Tool: claude-codeCommit: d0a6f962ac48e8387f4e589d098ebc0eaeccf7c5Authored: 4 days ago (Sat, 12 Sep 2026 17:31:59 +0000)Match: Claude-Session: https://claude.ai/code/session_017wP17UZAAo53fJNyHN8Whj
• AiGenerationPhrase
  Tool: claude-codeCommit: 6578ba42ea779851eaaf4b73452263daa3d74047Authored: 5 days ago (Sat, 12 Sep 2026 05:01:16 +0000)Match: Claude-Session: https://claude.ai/code/session_01UxDiPkoRCtgk6rMMNjPfdT
• AiGenerationPhrase
  Tool: claude-codeCommit: a8b3384c3aa6e3186dc741ace264de10e3e89df2Authored: 2 weeks ago (Thu, 3 Sep 2026 03:45:59 +0000)Match: Claude-Session: https://claude.ai/code/session_01TeuFQUkLJAMt4KTLvSF7hh
• AiCoAuthor
  Tool: claude-codeCommit: e402bd5424fd83c6cdbf6b4cf1b5574ab381076dAuthored: 1 month ago (Fri, 31 Jul 2026 03:49:24 +0000)Match: Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
• AiCoAuthor
  Tool: claude-codeCommit: 2ad4c58dcd3a67d985d48777ef31a949a15812a3Authored: 2 months ago (Mon, 6 Jul 2026 01:10:36 +0000)Match: Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
• AiCoAuthor
  Tool: claude-codeCommit: b6519e9839e44485e552be47aba6ff7dc337c8d8Authored: 2 months ago (Tue, 23 Jun 2026 05:03:38 +0000)Match: Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
• AiCoAuthor
  Tool: claude-codeCommit: c301b14cb3f0a1d35004c56226ac14bdb166a72fAuthored: 3 months ago (Wed, 17 Jun 2026 07:54:39 +0000)Match: Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
• AiCoAuthor
  Tool: claude-codeCommit: 5ac0aaeec09d10634532926564f37d8c5eea42afAuthored: 3 months ago (Tue, 2 Jun 2026 23:11:21 +0000)Match: Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
• AiCoAuthor
  Tool: claude-codeCommit: 22d37135ae52db97db665c6652fe497cdcce99adAuthored: 3 months ago (Mon, 25 May 2026 20:35:45 +0000)Match: Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
• AiCoAuthor
  Tool: claude-codeCommit: f3c0b68a0fac0d75c2f081f3a3902fd7263d7bd3Authored: 3 months ago (Tue, 19 May 2026 00:58:08 +0000)Match: Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
• AiCoAuthor
  Tool: claude-codeCommit: 7656e84c205b08b64ecc184e4a139473499094a7Authored: 6 months ago (Mon, 16 Feb 2026 20:44:29 +0000)Match: Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
• AiCoAuthor
  Tool: github-copilotCommit: f78a572a3cdb0851e19013a3197e64f30a2f2a91Authored: 8 months ago (Thu, 25 Dec 2025 22:22:23 +0000)Match: Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
• AiCoAuthor
  Tool: github-copilotCommit: a322717e0e396034ad37f04da04fcb4cad05627dAuthored: 8 months ago (Wed, 24 Dec 2025 04:46:19 +0000)Match: Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
• AiCoAuthor
  Tool: github-copilotCommit: 4889f2b51afc07e2bfa455caea42cb5e0c9af9caAuthored: 9 months ago (Tue, 2 Dec 2025 09:33:45 +0000)Match: Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>

source signals (18)

• File: AGENTS.mdTool: agents-md

  ## General guidelines
  
  - Never use the em dash "—". Use plain dash "-" instead.
  - On an explicit commit request: `git add` the changes, run `pnpm lint:staged`, then commit. It lints only staged TS/TSX/CSS/SCSS via eslint+stylelint and auto-fixes; typecheck runs in CI (`test.yml`), not locally. Do not stage or run it outside a commit request.
  
  ## Agent skills
  
  ### Issue tracker
  
  Issues live as markdown under `.scratch/<feature>/`. See `docs/agents/issue-tracker.md`.
  
  ### Triage labels
  
  Default vocabulary: 

• File: .claude/skills/caveman-commit/SKILL.mdTool: unspecified

  ---
  name: caveman-commit
  description: >
    Ultra-compressed commit message generator. Cuts noise from commit messages while preserving
    intent and reasoning. Conventional Commits format. Subject ≤50 chars, body only when "why"
    isn't obvious. Use when user says "write a commit", "commit message", "generate commit",
    "/commit", or invokes /caveman-commit. Auto-triggers when staging changes.
  ---
  
  Write commit messages terse and exact. Conventional Commits format. No fluff. Why over what.
  
  ## Rules
  
  **Subj

• File: .agents/skills/caveman-review/SKILL.mdTool: unspecified

  ---
  name: caveman-review
  description: >
    Ultra-compressed code review comments. Cuts noise from PR feedback while preserving
    the actionable signal. Each comment is one line: location, problem, fix. Use when user
    says "review this PR", "code review", "review the diff", "/review", or invokes
    /caveman-review. Auto-triggers when reviewing pull requests.
  ---
  
  Write code review comments terse and actionable. One line per finding. Location, problem, fix. No throat-clearing.
  
  ## Rules
  
  **Format:** `L<line>: 

• File: .agents/skills/caveman-commit/SKILL.mdTool: unspecified

  ---
  name: caveman-commit
  description: >
    Ultra-compressed commit message generator. Cuts noise from commit messages while preserving
    intent and reasoning. Conventional Commits format. Subject ≤50 chars, body only when "why"
    isn't obvious. Use when user says "write a commit", "commit message", "generate commit",
    "/commit", or invokes /caveman-commit. Auto-triggers when staging changes.
  ---
  
  Write commit messages terse and exact. Conventional Commits format. No fluff. Why over what.
  
  ## Rules
  
  **Subj

• File: .agents/skills/ponytail/SKILL.mdTool: unspecified

  ---
  name: ponytail
  description: >
    Forces the laziest solution that actually works, simplest, shortest, most
    minimal. Channels a senior dev who has seen everything: question whether the
    task needs to exist at all (YAGNI), reach for the standard library before
    custom code, native platform features before dependencies, one line before
    fifty. Supports intensity levels: lite, full (default), ultra. Use on ANY
    coding task: writing, adding, refactoring, fixing, reviewing, or designing
    code, and choos

• File: .claude/skills/caveman-review/SKILL.mdTool: unspecified

  ---
  name: caveman-review
  description: >
    Ultra-compressed code review comments. Cuts noise from PR feedback while preserving
    the actionable signal. Each comment is one line: location, problem, fix. Use when user
    says "review this PR", "code review", "review the diff", "/review", or invokes
    /caveman-review. Auto-triggers when reviewing pull requests.
  ---
  
  Write code review comments terse and actionable. One line per finding. Location, problem, fix. No throat-clearing.
  
  ## Rules
  
  **Format:** `L<line>: 

• File: .claude/skills/ponytail/SKILL.mdTool: unspecified

  ---
  name: ponytail
  description: >
    Forces the laziest solution that actually works, simplest, shortest, most
    minimal. Channels a senior dev who has seen everything: question whether the
    task needs to exist at all (YAGNI), reach for the standard library before
    custom code, native platform features before dependencies, one line before
    fifty. Supports intensity levels: lite, full (default), ultra. Use on ANY
    coding task: writing, adding, refactoring, fixing, reviewing, or designing
    code, and choos

• Dir: .agentsTool: unspecified
• Dir: .agentsTool: unspecified
• Dir: .agentsTool: unspecified
• Dir: .agentsTool: unspecified
• Dir: .agentsTool: unspecified
• File: CLAUDE.mdTool: claude-code

  AGENTS.md

• Dir: .claudeTool: claude-code
• Dir: .claudeTool: claude-code
• Dir: .claudeTool: claude-code
• Dir: .claudeTool: claude-code
• Dir: .claudeTool: claude-code

ignorefile signals (0)

none