.coderabbit.yaml

# yaml-language-server: $schema=https://coderabbit.ai/integrations/schema.v2.json
language: "en-US"
early_access: false
tone_instructions: "Only comment on issues introduced by this PR's changes. Do not flag pre-existing problems in moved, re-indented, or reformatted code."

reviews:
  profile: "assertive"
  request_changes_workflow: true
  high_level_summary: false
  poem: false
  review_status: false
  review_details: true
  commit_status: true
  # Without this, a review that never happened (rate limit, internal error)
  # still posts a green "CodeRabbit" commit status, so a throttled review is
  # indistinguishable from a clean one.
  fail_commit_status: true
  collapse_walkthrough: true
  changed_files_summary: false
  sequence_diagrams: false
  estimate_code_review_effort: false
  assess_linked_issues: false
  related_issues: false
  related_prs: false
  suggested_labels: false
  auto_apply_labels: false
  suggested_reviewers: false
  auto_assign_reviewers: false
  in_progress_fortune: false
  enable_prompt_for_ai_agents: true

  path_filters:
    - "!comfy_api_nodes/apis/**"
    - "!**/generated/*.pyi"
    - "!.ci/**"
    - "!script_examples/**"
    - "!**/__pycache__/**"
    - "!**/*.ipynb"
    - "!**/*.png"
    - "!**/*.bat"

  path_instructions:
    - path: "**"
      instructions: |
        IMPORTANT: Only comment on issues directly introduced by this PR's code changes.
        Treat AGENTS.md as mandatory repository policy, not optional style guidance.
        Flag PR changes that violate AGENTS.md even when the code is otherwise functional.
        In particular, enforce architecture boundaries, dtype/device/memory rules,
        interface contracts, import style, no unnecessary try/except blocks, no inline
        imports, no outbound internet paths in core ComfyUI, and narrow scoped fixes.
        Prefer direct findings over suggestions when a rule is violated. Only ignore
        AGENTS.md when it clearly conflicts with a newer explicit maintainer instruction
        in the PR.
        Do NOT flag pre-existing issues in code that was merely moved, re-indented,
        de-indented, or reformatted without logic changes. If code appears in the diff
        only due to whitespace or structural reformatting (e.g., removing a `with:` block),
        treat it as unchanged. Contributors should not feel obligated to address
        pre-existing issues outside the scope of their contribution.
    - path: "comfy/**"
      instructions: |
        Core ML/diffusion engine. Focus on:
        - Backward compatibility (breaking changes affect all custom nodes)
        - Memory management and GPU resource handling
        - Performance implications in hot paths
        - Thread safety for concurrent execution
    - path: "comfy_api_nodes/**"
      instructions: |
        Third-party API integration nodes. Focus on:
        - No hardcoded API keys or secrets
        - Proper error handling for API failures (timeouts, rate limits, auth errors)
        - Correct Pydantic model usage
        - Security of user data passed to external APIs
    - path: "comfy_extras/**"
      instructions: |
        Community-contributed extra nodes. Focus on:
        - Consistency with node patterns (INPUT_TYPES, RETURN_TYPES, FUNCTION, CATEGORY)
        - No breaking changes to existing node interfaces
    - path: "comfy_execution/**"
      instructions: |
        Execution engine (graph execution, caching, jobs). Focus on:
        - Caching correctness
        - Concurrent execution safety
        - Graph validation edge cases
    - path: "nodes.py"
      instructions: |
        Core node definitions (2500+ lines). Focus on:
        - Backward compatibility of NODE_CLASS_MAPPINGS
        - Consistency of INPUT_TYPES return format
    - path: "alembic_db/**"
      instructions: |
        Database migrations. Focus on:
        - Migration safety and rollback support
        - Data preservation during schema changes

  auto_review:
    enabled: true
    auto_incremental_review: true
    drafts: false
    ignore_title_keywords:
      - "WIP"
      - "DO NOT REVIEW"
      - "DO NOT MERGE"

  finishing_touches:
    docstrings:
      enabled: false
    unit_tests:
      enabled: false

  tools:
    ruff:
      enabled: false
    pylint:
      enabled: false
    flake8:
      enabled: false
    gitleaks:
      enabled: true
    shellcheck:
      enabled: false
    markdownlint:
      enabled: false
    yamllint:
      enabled: false
    languagetool:
      enabled: false
    github-checks:
      enabled: true
      timeout_ms: 90000
    ast-grep:
      essential_rules: true

chat:
  auto_reply: true

knowledge_base:
  opt_out: false
  code_guidelines:
    enabled: true
    filePatterns:
      - files: "AGENTS.md"
        applyTo: "**"
  linked_repositories:
    - repository: "Comfy-Org/ComfyUI_frontend"
      instructions: |
        Frontend consumer of Core APIs, routes, feature flags, asset behavior,
        node metadata, and package-version rollouts. Check for compatibility
        issues and changes that require coordinated rollout or merge ordering.
        Surface linked-repository findings only when they change a review
        conclusion or provide decision-relevant compatibility, ownership,
        rollout, or merge-order evidence. Continue using the repository as
        review context, but omit routine search inventories and no-match results
        that do not provide decision-relevant evidence.
    - repository: "Comfy-Org/comfy-kitchen"
      instructions: |
        Owns optimized inference operations and kernels used by ComfyUI. Check
        whether new model math or backend-specific operations duplicate an
        existing implementation or belong in comfy-kitchen instead of Core.
        Surface linked-repository findings only when they change a review
        conclusion or provide decision-relevant ownership, performance,
        compatibility, or rollout evidence. Continue using the repository as
        review context, but omit routine search inventories and no-match results
        that do not provide decision-relevant evidence.
  learnings:
    scope: "auto"