Skip to content
Open
Show file tree
Hide file tree
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
3 changes: 3 additions & 0 deletions AGENTS.md
Original file line number Diff line number Diff line change
Expand Up @@ -3,8 +3,11 @@
## Rules for AI Assistants
- **Use Makefile targets** instead of discovering build/test commands yourself.
- **Keep changes minimal.** Do not refactor, reorganize, or 'improve' code beyond what was explicitly requested.
- **Stay within scope.** Only modify what the task explicitly asks for. Do not make changes to files or components outside the stated scope, even if they seem related.
- **For CI/release workflows**, always use existing Makefile targets rather than reimplementing build logic in YAML.
- **Better tests.** Always try to add or improve tests(including integration, e2e) when modifying code.
- **Use typed constants.** In Go tests, always use typed constants (e.g., `batchv1.PodFailurePolicyActionCount`) instead of string literals (e.g., `"Count"`). This ensures compile-time checking and consistency with production code.
- **Keep docs in sync.** When modifying configuration or removing/adding features, update related documentation (e.g., README files) in the same PR.
- **Logging conventions.** Start log messages with capital letters and do not end with punctuation.
- **Commit messages.** Do not include PR links in commit messages.

Expand Down
3 changes: 3 additions & 0 deletions self-development/agentconfig.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -19,6 +19,7 @@ spec:
## Standards
- Do not create duplicate issues — check existing issues first with `gh issue list`
- Keep changes minimal and focused
- Stay within the stated scope of the task — do not modify files or components outside what was explicitly requested

## Project Conventions
- Use Makefile targets instead of discovering build/test commands yourself:
Expand All @@ -28,5 +29,7 @@ spec:
- `make test-integration` — run integration tests
- `make build` — build binary
- Always try to add or improve tests when modifying code
- In Go tests, use typed constants (e.g., `batchv1.PodFailurePolicyActionCount`) instead of string literals
- When modifying configuration or features, update related documentation (e.g., README) in the same PR
- Logging conventions: start log messages with capital letters and do not end with punctuation
- Commit messages: do not include PR links in commit messages
1 change: 1 addition & 0 deletions self-development/kelos-config-update.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -85,4 +85,5 @@ spec:
- Do not create a PR if no actionable changes are found — exit cleanly instead
- Do not create duplicate issues — check existing issues first with `gh issue list`
- Back every change with a specific reference to the PR review that motivated it
- When editing YAML files with `promptTemplate: |` (literal block scalars), ensure all content lines maintain the correct indentation level (typically 6 spaces for content inside a 4-space-indented key). Lines with insufficient indentation will exit the block scalar and break YAML parsing
pollInterval: 1m
1 change: 1 addition & 0 deletions self-development/kelos-self-update.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -82,5 +82,6 @@ spec:
- Strategic proposals, new use cases, integrations, CRD extensions → kelos-fake-strategist
- Documentation, CLI UX, error messages → kelos-fake-user
- Agent configuration based on PR reviews → kelos-config-update
- When editing or proposing changes to YAML files with `promptTemplate: |` (literal block scalars), ensure all content lines maintain the correct indentation level (typically 6 spaces for content inside a 4-space-indented key). Lines with insufficient indentation will exit the block scalar and break YAML parsing
- If after review you find nothing actionable to improve, exit without creating an issue or PR. Not every run needs to produce output
pollInterval: 1m
Loading