knowledge-curation
Scannednpx machina-cli add skill a5c-ai/babysitter/knowledge-curation --openclawKnowledge Curation
Overview
Two-phase knowledge management: prime context before work starts, and extract learnings after work completes. Knowledge persists in JSONL files for cross-session continuity.
When to Use
- Before starting any work (prime mode)
- After completing work, BEFORE PR creation (reflect mode)
- When recovering from context loss (recovery priming)
Knowledge Categories
| Category | File | Content |
|---|---|---|
| Critical Rules | facts.jsonl | MUST FOLLOW constraints |
| Gotchas | gotchas.jsonl | Common pitfalls |
| Patterns | patterns.jsonl | Codebase best practices |
| Decisions | decisions.jsonl | Architectural choices with rationale |
| Anti-Patterns | anti-patterns.jsonl | What NOT to do |
| Codebase Facts | codebase-facts.jsonl | Structural information |
| API Behaviors | api-behaviors.jsonl | Undocumented quirks |
Process
Prime Mode
- Load knowledge base files for work type
- Surface MUST FOLLOW rules first
- Present GOTCHAS and PATTERNS
- Load relevant DECISIONS
Reflect Mode
- Extract patterns from completed work
- Identify gotchas from review failures
- Record architectural decisions with rationale
- Persist to .beads/knowledge/
Tool Use
Invoke via babysitter process: methodologies/metaswarm/metaswarm-knowledge-cycle
Source
git clone https://github.com/a5c-ai/babysitter/blob/main/plugins/babysitter/skills/babysit/process/methodologies/metaswarm/skills/knowledge-curation/SKILL.mdView on GitHub Overview
Knowledge Curation is a two-phase process: prime context before work and reflect after completion to capture patterns, gotchas, and decisions into a centralized knowledge base. The knowledge persists in JSONL files to ensure cross-session continuity.
How This Skill Works
In Prime Mode, the system loads knowledge base files for the current work type, surfaces MUST FOLLOW rules first, and presents GOTCHAS and PATTERNS, then loads relevant DECISIONS. In Reflect Mode, it extracts patterns from completed work, identifies gotchas from review failures, records architectural decisions with rationale, and persists updates to .beads/knowledge/.
When to Use It
- Before starting any work (prime mode)
- After completing work, BEFORE PR creation (reflect mode)
- When recovering from context loss (recovery priming)
- During onboarding or knowledge transfer to a new teammate
- After major codebase changes to refresh patterns, rules, and decisions
Quick Start
- Step 1: Load knowledge base files for the current work type
- Step 2: Step into Prime Mode to surface MUST FOLLOW, GOTCHAS, PATTERNS, and load DECISIONS
- Step 3: After completion, run Reflect Mode and persist updates to .beads/knowledge/
Best Practices
- Load knowledge base files for the current work type at the start of a task
- Always surface MUST FOLLOW rules first, then GOTCHAS and PATTERNS
- Load and review relevant DECISIONS before proceeding
- Persist learnings to .beads/knowledge/ after Reflect Mode
- Use recovery priming to prevent context loss during long-running tasks
Example Use Cases
- Prime context before a feature rewrite, then reflect post-implementation to capture decisions and gotchas
- After submitting a PR, perform reflect steps to codify learnings and architectural rationales
- Run recovery priming after a context loss to restore critical constraints and patterns
- Onboard a new teammate by pre-loading knowledge base with critical rules and patterns
- After a large codebase refactor, refresh patterns, anti-patterns, and decisions with rationales
