skills.rest
MileniumTick
MileniumTickCOMMUNITY
Explore all skills by this author →
GitHub

skill-idea-miner

Mine Claude Code session logs to generate candidate skill ideas and backlog YAML.

Updated Mar 25, 2026
Search Tags:#automation#llm#backlog#deduplication#session-logs#skill-idea#idea-mining
One-click install
npx skills add https://github.com/MileniumTick/skills --skill skill-idea-miner-mileniumtick
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-idea-miner
Source: https://github.com/MileniumTick/skills/tree/main/skills/skill-idea-miner
Command: npx skills add https://github.com/MileniumTick/skills --skill skill-idea-miner-mileniumtick

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pyyaml, and includes scripts (resource) and references (resource) components.

What problem does it solve?

Mine Claude Code session logs to surface actionable skill ideas and backlog them, reducing manual discovery time and enabling rapid iteration on new capabilities.

Core Features & Use Cases

  • Session mining: extracts user interactions and tool usage from recent Claude sessions.
  • Deduplication: compares against existing SKILL.md frontmatter and backlog ideas to avoid duplicates.
  • Backlog output: writes a backlog YAML with ideas, statuses, and timestamps for governance.
  • Optional LLMed scoring: supports scoring ideas for novelty, feasibility, and trading value to populate backlog.

Quick Start

Run the skill-idea-miner pipeline to generate raw_candidates.yaml and backlog entries from the latest Claude session logs.

Frequently Asked Questions about skill-idea-miner

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I extract new skill ideas from Claude Code session logs?▼

Session logs are mined using deterministic signal detection to extract user interactions and tool usage. This process generates raw_candidates.yaml and populates an idea backlog, surfacing actionable skill concepts while reducing manual discovery time.

How does deduplication work when mining ideas from existing SKILL.md files?▼

Deduplication compares newly mined session signals against existing SKILL.md frontmatter and the current idea backlog. This frontmatter-based matching prevents duplicate concepts from populating the logs/.skill_generation_backlog.yaml output file.

Can I score mined backlog ideas for novelty and feasibility?▼

Yes, the pipeline supports optional LLMed scoring to evaluate mined ideas for novelty, feasibility, and trading value. These scores populate the backlog YAML alongside idea statuses and timestamps for governance.

What is the best way to automate weekly skill-generation pipelines across multiple projects?▼

Running an automated pipeline that mines recent Claude sessions across multiple projects handles weekly skill-generation. It scores ideas for novelty and feasibility, integrating them with existing SKILL.md definitions to output a structured YAML backlog.

Do I need pyyaml to generate the raw_candidates.yaml output?▼

Yes, pyyaml is required because the pipeline outputs raw_candidates.yaml and writes to logs/.skill_generation_backlog.yaml. This dependency handles parsing existing SKILL.md frontmatter and serializing the structured backlog files.

Related Skills

Explore More Software Engineering Skills→
Driving-TeacherDriving-Teacher
1

session-analyzer

Analyze Claude Code session logs to verify skill execution against SKILL.md.

Official
Advanced
bitwardenbitwarden
129

extracting-session-data

Extract structured data from Claude Code session logs via command-line scripts.

Official
Intermediate
Fabio29TFabio29T
2

skill-idea-miner

Mine Claude Code session logs to extract, score, and backlog skill ideas.

Community
Advanced
SensLiaoSensLiao

gsd-capture

Capture ideas, tasks, and notes into the GSD system using YAML frontmatter and Markdown.

Community
Intermediate