Generate concept_summary.md from an idea specification for planning and traceability.
npx skills add https://github.com/Darkpunkki/v-forge --skill summarize-darkpunkki
Please help me install this Agent Skill. Skill: Summarize Source: https://github.com/Darkpunkki/v-forge/tree/main/.claude/skills/concept-summarizer Command: npx skills add https://github.com/Darkpunkki/v-forge --skill summarize-darkpunkki
Converts a raw idea/spec into a stable invariant concept_summary.md that serves as the planning anchor for ongoing work, runs, and updates to the latest summary.
Invoke the concept-summarizer with an IDEA_ID to generate a new concept_summary and update the latest anchor.
High-intent search queries and answers about installing and using this skill.
To create a stable concept summary from a raw idea specification, the Skill generates an invariant semantic anchor concept_summary.md. This file serves as a persistent planning anchor during ongoing exploration and development runs.
Maintaining planning traceability across runs is achieved by generating per-run concept_summary.md files and logging activity to run_log.md. The Skill also updates a latest concept_summary.md to reflect the current idea understanding for quick reference.
A concept summary anchor works by providing a persistent semantic reference point derived from an idea specification. It ensures that ongoing work and updates remain aligned with the original concept, updating the latest summary for quick reference.
You do not need a normalized idea file to generate a concept summary. The Skill requires a baseline idea spec at inputs/idea.md, but optionally uses a normalized input at latest/idea_normalized.md and a concept_config.md to guide the summary output.
You can audit changes when updating the latest concept summary because the Skill produces a run_log entry and manifest updates. It maintains a per-idea manifest with status and provenance, and logs activity to run_log.md for full auditability.
A limitation of using a single invariant concept summary is that it applies to one specific IDEA_ID, writing outputs to dedicated run and latest directories. It relies on the initial idea specification and optional normalized inputs, which may constrain dynamic scope changes.
smallnestExtract and analyze content from .docx documents for semantic search and QA retrieval.
kyldvsConvert vague project ideas into Claude-executable PLAN.md files with atomic tasks and checkpoints.
jeffrigbySynchronize project documentation with code changes across CLAUDE.md, README.md, CHANGELOG.md, and /docs.
bobvarkeySearch document content by meaning and summarize relevant passages.
