craft-promote
Skill0 starsChanged 56 days agoMITView on GitHub
Promote a universal engineering-ledger lesson (LL) into a new Craft-native skill. Use when LL scope is universal, promotion candidate is yes, and the pattern should apply across projects. Checks for duplicates against Addy pack before creating skill.
In this repository5 other files it gives its agents.
- AGENTS.mdAGENTS.md
- craft-adoptSkill
- engineering-ledgerSkill
- product-managementSkill
- using-craftSkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 1 is distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 50
If duplicate found in Addy → stop.
- Distinctive: line 89
Add to README.md skills table
- Very specific: line 90
Add to craft.manifest.yaml under craft_native
- Very specific: line 91
Add router row in using-craft/SKILL.md if phase-specific
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Craft Promote
Convert universal ledger lessons into reusable Craft skills. This creates new Craft-native workflows — never copies Addy skill bodies.
Criteria: promotion-criteria.md
When to use
- LL entry has
Scope: universalandPromotion candidate: yes - Same pattern appeared in ≥2 projects OR user explicitly requests promotion
- Pattern is not covered by existing Addy skill
When NOT to use
Scope: projectlessons (stay in project ledger)- Duplicate of Addy skill (update
using-craftrouter to link Addy instead) - One-off incident with no general mechanism
Workflow
1. Extract from LL entry
From project docs/engineering-ledger/lessons.md, gather:
- Title → skill name (kebab-case)
- General lesson → skill purpose
- Root cause / fix → workflow steps
- Evidence → verification gate
- What went wrong → anti-patterns table
2. Dedup check
bash skills/craft-promote/scripts/check-duplicate.sh "<proposed-skill-name>"
bash skills/craft-promote/scripts/check-duplicate.sh "<key phrase from general lesson>"If duplicate found in Addy → stop. Add router row in using-craft pointing to existing skill; mark LL promoted-via-link: <skill-name>.
3. Scaffold skill
Create skills/<kebab-name>/SKILL.md:
---
name: <kebab-name>
description: >-
<one line — triggers for agent>
---
# <Title>
## When to use
...
## When NOT to use
...
## Workflow
1. ...
## Verification
- [ ] ...
## Anti-patterns
| Don't | Do instead |
...
## Originating lessons
- LL-### from <project/repo> (YYYY-MM-DD)Optional: references/ for long examples.
4. Register skill
- Add to README.md skills table
- Add to
craft.manifest.yamlundercraft_native - Add router row in
using-craft/SKILL.mdif phase-specific
5. Update source LL
In project ledger, set:
- **Status:** promoted
- **Promoted to:** craft/skills/<kebab-name>Verification gate
- SKILL.md has frontmatter, triggers, workflow, verification, anti-patterns
- Dedup script run; no Addy overlap (or linked instead)
- Originating LL ids cited
- README + manifest updated
- No copied text from upstream packs
Anti-patterns
| Don't | Do instead |
| Fork Addy skill into Craft | Reference Addy in router |
| Promote without verification gate | Write testable exit criteria |
| Delete original LL | Mark promoted; keep history |
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

