hyper-docs-sync

Synchronize project documentation with recent code changes via diff scopes.

3|Updated May 10, 2026
One-click install
npx skills add https://github.com/zeikar/hyperclaude --skill hyper-docs-sync
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: hyper-docs-sync
Source: https://github.com/zeikar/hyperclaude/tree/main/skills/hyper-docs-sync
Command: npx skills add https://github.com/zeikar/hyperclaude --skill hyper-docs-sync

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill prevents docs from going stale by syncing documentation to match recent code changes in APIs, schemas, CLI flags, and architecture.

Core Features & Use Cases

  • Detects code changes for a targeted doc update: Computes the changed file set and diffs for a range of invocation scopes (empty since main, uncommitted, specific commit SHA, or a ref range).
  • Maps code changes to documentation via a CLAUDE mapping table: Reads CLAUDE.md and/or AGENTS.md to find a Code-to-Docs mapping table and uses it for precise routing.
  • Uses safe heuristics with confidence reporting: Applies name-similarity heuristics only when appropriate, and skips low-confidence matches while reporting possible candidates.
  • Creates or updates affected docs using a documenter agent: Dispatches the documenter agent per affected target doc (create mode if the doc file is missing, otherwise update mode) and aggregates multiple code diffs to avoid redundant dispatches.

Quick Start

Run /hyperclaude:hyper-docs-sync after your implementation work to automatically update the impacted documentation files to reflect the current code state.

Frequently Asked Questions about hyper-docs-sync

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

FAQPage Schema
How do I keep API documentation synced with recent code changes?

Documentation sync prevents outdated claims by computing changed file sets from git diffs, mapping them via CLAUDE.md Code-to-Docs tables, and dispatching a documenter agent to update affected API, schema, and CLI flag documentation automatically.

Can I update documentation for a specific git commit SHA or uncommitted changes?

Documentation sync supports multiple git diff scopes including specific commit SHAs, uncommitted working-tree changes, main-branch diffs, and ref ranges, mapping each changed file set to the relevant target docs for batched updates.

How does code-to-doc mapping work when there is no CLAUDE.md mapping table?

Code-to-doc mapping uses guarded name-similarity heuristics as a fallback when CLAUDE.md or AGENTS.md tables are absent, skipping low-confidence matches while reporting possible candidate docs for manual review.

What is the best way to automate architecture and schema documentation updates?

Automating schema and architecture documentation updates is best handled by dispatching a documenter agent per affected target doc, aggregating multiple code diffs to avoid redundant dispatches while applying changes based on precise mapping rationale.

Do I need a CLAUDE.md file to use this documentation sync tool?

A CLAUDE.md file is not required; the Skill parses CLAUDE.md or AGENTS.md Code-to-Docs tables when available for precise routing, but safely falls back to guarded name-similarity heuristics when they are missing.