Distill kb
Knowledge Distiller: Consolidates task entries into long-term skill files. Decoupled for open-source.From its SKILL.md
npx -y skills add Allen930311/auto-skill-core --skill distill-kbAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
1.4 KB, 304 tokens by cl100k_base, as published. Nobody here has run it
Distill-KB (Core)
Overview
Distillation is the process of moving "raw task logs" from the Pending Zone of a skill file into structured Permanent Knowledge (Success/Failure cases).
Workflow
Step 1: Scan Targets
- Load the experience index from
{vault}/experience/_index.json. - Identify skill files with content in the
## 📝 Pending Zone.
Step 2: Categorization
Route entries based on their result:
- Success/Partial Success: Route to
legacy-{id}.md. - Failure: Route to
skill-{id}-worst-practice.md.
Step 3: Refinement
- Extract core commands, SOPs, and "Pitfalls to Avoid".
- Update the main
skill-{id}.mdwith refined best practices.
Step 4: Indexing
- Update the Case Index in the main skill file with pointers to the full legacy records.
Technical Implementation (Python snippet reference)
The system uses the Config loader to resolve paths:
from scripts.experience_api import Config
config = Config.load()
experience_base = config.get_path("vault") / "experience"
Quality Checklist
- Refined logic reduces line count of the original entry by >50%.
- No hardcoded private paths in distilled content.
- All keywords and metadata updated in
_index.json.
What ships with it
Read from the repository
Just SKILL.md. No reference files, no scripts.