菜单
Provides a structured logic for capturing and routing agentic learnings to prevent knowledge pollution.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Automate the initial scaffolding of a KCC "direct" resource, including CRD types and generation scripts. Use this when starting a new "direct" implementation for a GCP resource.
Creates or updates the _identity.go and _reference.go files for a Config Connector resource, ensuring they follow the canonical gcpurls.Template pattern. Use this when you need to make sure the identity and reference is up to date for a KCC resource or when implementing IdentityV2 and refs.Ref for a resource.
Standards and workflows for creating direct KRM Go types for an existing CRD while maintaining strict schema compatibility.
Comprehensive review guide and criteria for PRs adding or modifying features in legacy resources (TF patches, DCL patches, CRD changes, MockGCP alignment, and Direct Controller alignment).
Implement the controller, mappers, and fuzzer for a direct KCC resource, ensuring package isolation and CI compliance. Use this when implementing the main reconciliation logic for a "direct" resource.
Guides modifying, patching, and updating resources and fields managed by the Terraform/legacy controller, including patching the local/vendored Terraform Google Beta provider.
| name | kcc-agentic-journaler |
| description | Provides a structured logic for capturing and routing agentic learnings to prevent knowledge pollution. |
As an agent, you must record "tribal knowledge" and technical breakthroughs. This skill ensures that your learnings are stored in the most contextually relevant location.
When you have a new learning, follow this hierarchy to choose the destination:
Scenario: You found a better way to do something that applies to all KCC resources (e.g., "A more efficient way to resolve Project IDs").
Destination: Update the relevant Skill's SKILL.md file directly.
Scenario: You found a trick for a specific implementation area, or a quirk unique to a specific GCP service.
Destination: ALWAYS create or append to .gemini/journals/<service_name>.md.
CRITICAL: Do NOT append to any journal.md inside a .gemini/skills/ directory. Multiple agents working in parallel on different services will cause merge conflicts if they all write to a centralized skill journal. By keeping journals scoped to the <service_name>, we eliminate cross-service merge conflicts.
Use this format for your entries in .gemini/journals/<service_name>.md:
### [YYYY-MM-DD] <Brief Title>
- **Context**: <What were you implementing? Link to Kind/PR>
- **Problem**: <What was the unexpected behavior or hurdle?>
- **Solution**: <How did you solve it? Provide code snippets if relevant.>
- **Impact**: <Why should the next agent care about this?>
In your final turn of the task, you MUST state which knowledge was captured and provide the file path.