Skip to content
Merged
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
62 changes: 60 additions & 2 deletions .coderabbit.yaml
Original file line number Diff line number Diff line change
@@ -1,7 +1,9 @@
# yaml-language-server: $schema=https://coderabbit.ai/integrations/schema.v2.json
inheritance: true
language: en-US
early_access: false
# Lifted from the Org-UI Web dashboard (2026-05-12) so this YAML is the
# canonical source of truth. Org-UI editor should be cleared after this lands.
early_access: true
tone_instructions: >
Concise, technical, no filler. Focus on correctness, security, performance,
maintainability. Cite file paths and line numbers.
Expand All @@ -24,10 +26,19 @@ reviews:
auto_apply_labels: false
suggested_reviewers: false

# Lifted from Org-UI Web dashboard (2026-05-12).
high_level_summary_in_walkthrough: true
review_details: true
fail_commit_status: true
estimate_code_review_effort: false
in_progress_fortune: false

auto_review:
enabled: true
auto_incremental_review: true
drafts: false
# Lifted from Org-UI (2026-05-12). 0 = never auto-pause after N reviewed commits.
auto_pause_after_reviewed_commits: 0
base_branches:
- current
- sagitta
Expand All @@ -38,6 +49,14 @@ reviews:
- WIP
- DO NOT MERGE
- Bump
# Skip auto-review for PRs authored by these accounts. Manual
# `@coderabbitai review` still works if a human asks for one. The list
# targets bots that open mechanical PRs whose content was already
# reviewed upstream (e.g. Mergify cherry-picks the merge commit of an
# already-reviewed source PR, so re-reviewing the backport adds no
# signal and burns rate limit).
ignore_usernames:
- "mergify[bot]"

path_filters:
- "!**/*.lock"
Expand Down Expand Up @@ -65,9 +84,16 @@ reviews:
- "!**/.terragrunt-cache/**"

finishing_touches:
# Both docstring and unit-test generation are disabled at the org-wide
# level; per-repo `.coderabbit.yaml` files re-enable either feature
# where the stack (language, testing framework, doc conventions)
# makes it a good fit. Fleet-wide generation is too noisy because
# the conventions differ across repos.
docstrings:
enabled: true
enabled: false
unit_tests:
enabled: false
simplify:
enabled: true

tools:
Expand All @@ -79,12 +105,44 @@ reviews:
yamllint: { enabled: true }
actionlint: { enabled: true }
ast-grep: { essential_rules: true }
# Linters disabled fleet-wide because they don't match our stack
# (Swift/Ruby/Rails/Ember/Shopify/CircleCI/PMD). Lifted from Org-UI
# Web dashboard (2026-05-12).
swiftlint: { enabled: false }
detekt: { enabled: false }
rubocop: { enabled: false }
regal: { enabled: false }
pmd: { enabled: false }
circleci: { enabled: false }
shopifyThemeCheck: { enabled: false }
brakeman: { enabled: false }
blinter: { enabled: false }
smartyLint: { enabled: false }
emberTemplateLint: { enabled: false }

chat:
auto_reply: true
# Lifted from Org-UI Web dashboard (2026-05-12). Restricts CodeRabbit
# chat to org members; non-members receive a deny message.
allow_non_org_members: false
integrations:
jira: { usage: disabled }
linear: { usage: disabled }

knowledge_base:
opt_out: false
learnings: { scope: auto }
issues: { scope: auto }
pull_requests: { scope: auto }
# Lifted from Org-UI Web dashboard (2026-05-12).
jira: { usage: disabled }
linear: { usage: disabled }
mcp: { usage: enabled }
# Linked repositories provide cross-repo context to CodeRabbit. Org-specific
# — vyos org consumes Ansible context because the `vyos.vyos` Ansible
# collection lives here and references upstream ansible/ansible patterns.
linked_repositories:
- repository: ansible/ansible
instructions: >
This is the main repo for Ansible; our current repo is a collection
for Ansible.