.claude/voyager/generated_skills_index.json — index of generated skills
Safe Two-Step Loop
Always follow this workflow:
Step 1: Propose Candidates
Analyze patterns and present 3-5 skill candidates to the user:
voyager factory propose
Creates skill_proposals.json with candidates
Present each with name, description, triggers, and rationale
Wait for user selection before proceeding
Step 2: Scaffold User's Choice
Only scaffold the skill the user explicitly selects:
# List available proposals
voyager factory list
# Scaffold by name (after user selects)
voyager factory scaffold --name <selected-skill>
# Or by index
voyager factory scaffold --index <n>
Important: Never scaffold multiple skills automatically. Let the user pick.
Common Queries
"What skills could I create?"
Read .claude/voyager/brain.json to understand recent patterns
Run voyager factory propose --dry-run to see proposals
Present the top 3 proposals with rationale
"Turn this workflow into a skill"
Ask user to describe the workflow
Create a manual proposal based on their description
Run scaffold_skill.py with the proposal
Review and refine the generated SKILL.md
"Generate a skill for X"
Analyze what "X" involves (tools needed, triggers, workflow)
Create or find a matching proposal
Scaffold the skill
Add it to the generated skills index
Files
File
Purpose
schemas/skill_proposal.schema.json
Validates skill proposals
prompts/propose_skills.prompt.md
LLM prompt for proposing skills
prompts/scaffold_skill.prompt.md
LLM prompt for generating SKILL.md
reference.md
Technical reference
Guardrails
Never overwrite existing skills without explicit confirmation
Validate YAML frontmatter before scaffolding
Check duplicates against existing generated skills
Keep SKILL.md concise (< 100 lines, point to references)
See reference.md for full implementation details.
Converted and distributed by TomeVault — claim your Tome and manage your conversions.
1---2name: zenbase-ai-code-voyager-skill-factory3description: Skill Factory4---56# Skill Factory78Proposes and scaffolds new skills from observed patterns in transcripts and brain state.910## Outputs1112- `.claude/skills/generated/<skill-name>/` — scaffolded skill folders13- `.claude/voyager/skill_proposals.json` — latest skill proposals14- `.claude/voyager/generated_skills_index.json` — index of generated skills1516## Safe Two-Step Loop1718**Always follow this workflow:**1920### Step 1: Propose Candidates2122Analyze patterns and present **3-5 skill candidates** to the user:2324```bash25voyager factory propose26```2728- Creates `skill_proposals.json` with candidates29- Present each with name, description, triggers, and rationale30- **Wait for user selection** before proceeding3132### Step 2: Scaffold User's Choice3334Only scaffold the skill the user explicitly selects:3536```bash37# List available proposals38voyager factory list3940# Scaffold by name (after user selects)41voyager factory scaffold --name <selected-skill>4243# Or by index44voyager factory scaffold --index <n>45```4647**Important:** Never scaffold multiple skills automatically. Let the user pick.4849## Common Queries5051### "What skills could I create?"52531. Read `.claude/voyager/brain.json` to understand recent patterns542. Run `voyager factory propose --dry-run` to see proposals553. Present the top 3 proposals with rationale5657### "Turn this workflow into a skill"58591. Ask user to describe the workflow602. Create a manual proposal based on their description613. Run scaffold_skill.py with the proposal624. Review and refine the generated SKILL.md6364### "Generate a skill for X"65661. Analyze what "X" involves (tools needed, triggers, workflow)672. Create or find a matching proposal683. Scaffold the skill694. Add it to the generated skills index7071## Files7273| File | Purpose |74|------|---------|75| `schemas/skill_proposal.schema.json` | Validates skill proposals |76| `prompts/propose_skills.prompt.md` | LLM prompt for proposing skills |77| `prompts/scaffold_skill.prompt.md` | LLM prompt for generating SKILL.md |78| `reference.md` | Technical reference |7980## Guardrails8182- **Never overwrite** existing skills without explicit confirmation83- **Validate** YAML frontmatter before scaffolding84- **Check duplicates** against existing generated skills85- **Keep SKILL.md concise** (< 100 lines, point to references)8687See `reference.md` for full implementation details.8889---90> Converted and distributed by [TomeVault](https://tomevault.io/claim/zenbase-ai) — claim your Tome and manage your conversions.91<!-- tomevault:4.0:skill_md:2026-04-11 -->
Run npx skillmds@latest add tomevault-io/zenbase-ai-code-voyager-skill-factory in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.
Skill Factory It is listed under Coding & Dev Tools on SkillMD.
This skill has not completed SkillMD's automated safety review yet. Independent scanners report: SkillSpector: PASS, Skill Scanner: PASS. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.
This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.
Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.
tomevault-io (@tomevault-io) published this skill. Their other Agent Skills are listed on their SkillMD profile.
tomevault-io