skill-idea-miner

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

2|Updated Mar 4, 2026
One-click install
npx skills add https://github.com/Fabio29T/Trading-Skills --skill skill-idea-miner
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-idea-miner
Source: https://github.com/Fabio29T/Trading-Skills/tree/main/skills/skill-idea-miner
Command: npx skills add https://github.com/Fabio29T/Trading-Skills --skill skill-idea-miner

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires python3, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill automates the process of identifying potential new Claude skills by analyzing session logs, reducing the manual effort required for idea generation and backlog maintenance.

Core Features & Use Cases

  • Log Analysis: Scans Claude Code session logs for patterns indicating user needs or automation opportunities.
  • Signal Detection: Identifies skill usage, errors, repetitive tasks, and direct automation requests.
  • Idea Abstraction: Uses an LLM to abstract raw signals into concrete skill ideas.
  • Deduplication & Scoring: Compares new ideas against existing skills and backlog items, scoring them for novelty, feasibility, and trading value.
  • Use Case: Run this Skill weekly to automatically populate your skill development backlog with data-backed ideas derived from actual user interactions.

Quick Start

Run the weekly skill generation pipeline to mine session logs for new skill ideas.

Frequently Asked Questions about skill-idea-miner

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

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

You can discover new skill ideas by mining session logs for deterministic signals like skill usage frequency, error patterns, repetitive tool sequences, and automation request keywords, then abstracting them with an LLM.

What is the best way to automate backlog management for a skill repository?

Automating backlog management involves analyzing session logs for repetitive tasks and errors, deduplicating extracted ideas against existing skills using Jaccard similarity, and scoring them for novelty, feasibility, and trading value.

How does LLM-based idea abstraction work for log analysis?

LLM-based idea abstraction works by parsing raw session log signals and using a large language model to convert repetitive tool sequences and error patterns into concrete, actionable skill development candidates.

Do I need Python3 to run the skill idea mining pipeline?

Yes, you need Python3 installed because the Skill requires Python scripts to execute core tasks like parsing session logs, interacting with the LLM, and calculating Jaccard similarity for deduplication.

How are new skill ideas deduplicated against an existing backlog?

New skill ideas are deduplicated by comparing them against existing skills and backlog items using Jaccard similarity, ensuring only novel ideas are retained for scoring on feasibility and trading value.

What signals indicate an automation opportunity in session logs?

Signals indicating automation opportunities include high skill usage frequency, recurring error patterns, repetitive tool sequences, and direct automation request keywords found within the Claude Code session logs.