Extract epics

Extract 2–12 epics from idea documents using concept_summary.md as the semantic anchor.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/Darkpunkki/v-forge --skill extract-epics
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: Extract epics
Source: https://github.com/Darkpunkki/v-forge/tree/main/.claude/skills/epic-extractor
Command: npx skills add https://github.com/Darkpunkki/v-forge --skill extract-epics

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Convert idea-level context into a well-scoped backlog of epics that can guide product planning and execution.

Core Features & Use Cases

  • Generate 2–12 epics from an idea using concept_summary.md as the semantic anchor.
  • Write backlog artifacts to runs/<RUN_ID>/outputs/epics_backlog.md and update ideas/<IDEA_ID>/latest/epics_backlog.md, run_log.md, and manifest.md.
  • Update the central idea governance documentation and allow tracing of provenance for epics.

Quick Start

Run /epic-extractor <IDEA_ID> to generate a structured backlog of epics from your idea.

Frequently Asked Questions about Extract epics

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

FAQPage Schema
How do I generate a structured backlog of epics from a project idea?

Generate a structured backlog of epics by running the epic extractor on an idea document, which reads concept_summary.md to derive 2–12 well-scoped epics. The results are written to epics_backlog.md with per-epic metadata including id, title, priority, and dependencies.

What is the best way to convert a concept summary into non-overlapping project epics?

The best way to convert a concept summary into project epics is using a semantic anchor file like concept_summary.md. This ensures derived epics have clear scope and non-overlapping boundaries, producing a governed backlog ready for execution.

How does idea governance track the provenance of generated epics?

Idea governance tracks provenance by updating idea governance documentation and run logs. It writes epics to ideas/<IDEA_ID>/latest/epics_backlog.md and records metadata like last_updated and last_run_id to maintain traceability across runs.

Can I extract epics if I only have a basic idea.md file without normalized inputs?

Yes, you can extract epics with only idea.md. The process reads concept_summary.md and idea.md to derive epics, while optionally using idea_normalized.md if present to enhance the backlog structure and scope definition.

How many epics can I expect to get when extracting from a single idea document?

Extracting from a single idea document produces between 2 and 12 epics. Each epic includes clear scope boundaries and metadata such as id, title, status, release_target, priority, and depends_on to guide execution.