Menu
Use when the user says: "create a skill", "write a skill", "new skill", or wants to author a new SKILL.md. Guided skill creation following Claudefiles conventions.
Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
Basé sur la classification professionnelle SOC
'Use this when the users says: "review existing feature","make this change"
Use when the user says: "extract learnings", "save this for next time", "remember this pattern", "consolidate memories", "dream", "clean up memories".
Use when the user asks about Claude token usage, wants to see how much they are spending on Claude, understand cache hit rates, review Claude Code workflow patterns, or get cost optimization recommendations.
Use when the user asks to recall, search, or continue past conversations. Triggers on "what did we discuss", "continue where we left off", "remember when", "as I mentioned", "you suggested", "we decided", "search my conversations", "find the conversation where", "what did we work on". Also triggers on implicit signals like past-tense references ("the bug we fixed"), possessives without context ("my project"), or assumptive questions ("do you remember").
Generate a commit message
Manage conversations and task context
| name | mine.write-skill |
| description | Use when the user says: "create a skill", "write a skill", "new skill", or wants to author a new SKILL.md. Guided skill creation following Claudefiles conventions. |
| user-invocable | true |
Guided creation of a new skill for this repo. Gathers requirements, drafts the skill, validates against a quality checklist, and writes it to the correct location.
$ARGUMENTS — optional skill name or description. If provided, use as starting context.
Ask these one at a time. Skip any already answered by $ARGUMENTS.
AskUserQuestion:
question: "What task should this skill handle? Describe the problem it solves."
header: "Purpose"
AskUserQuestion:
question: "What trigger phrases should invoke it? (e.g., when the user says 'audit the codebase')"
header: "Triggers"
AskUserQuestion:
question: "What does the output look like? (e.g., a file, a report, a question, an action)"
header: "Output"
AskUserQuestion:
question: "Should this be user-invocable (slash command) or a reference skill (loaded by other skills only)?"
header: "Invocability"
multiSelect: false
options:
- label: "User-invocable"
description: "Can be called directly via /mine.<name>"
- label: "Reference only"
description: "Loaded by other skills, not directly callable"
Then explore the codebase for related skills, patterns, or existing work that should inform the new skill's design.
Derive from the purpose: mine.<kebab-case-name>. Max 30 chars total. Check that skills/mine.<name>/ doesn't already exist.
Write to skills/mine.<name>/SKILL.md using the template in REFERENCE.md. See REFERENCE.md for conventions, AskUserQuestion constraints (header ≤12 chars, max 4 options), and size guidance.
Validate the drafted skill against the quality checklist in REFERENCE.md (full criteria there). Quick list:
Report results. Fix any failures before presenting to the user.
After the user approves the skill:
rules/common/capabilities.md using the trigger phrases from Phase 1README.md (alphabetical order) and update the skill count in the section header./install.sh to create the symlink