source-unifier_skill

This skill intelligently unifies documentation from multiple sources—docs, GitHub, and PDFs—and reports conflicts for consistent, accurate AI-assisted
  • TypeScript

92

GitHub Stars

1

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill jmagly/aiwg --skill source-unifier

  • SKILL.md7.7 KB

Overview

This skill intelligently merges multiple documentation sources (web docs, GitHub repos, PDFs) into a single unified reference while detecting and reporting conflicts between documentation and implementation. It produces a merged output with clear conflict annotations and a machine-readable conflict summary to support review and remediation. Use it to create complete, auditable skill coverage combining code and docs.

How this skill works

The unifier verifies source accessibility and type, extracts API signatures and content from each source, and compares items to detect missing items, signature mismatches, and description differences. It highlights conflicts by severity, saves checkpoints during processing, and emits a unified output layout with both human-readable conflict reports and structured JSON for automation. The tool supports rule-based, AI-enhanced, and manual merge modes and refuses to silently resolve unclear conflicts.

When to use it

  • Combining product docs, repository code, and PDFs into a single authoritative reference
  • Preparing a skill or library manifest that must reflect both documentation and implementation
  • Auditing API surface to find undocumented or misdocumented features
  • Automating release notes or migration work where docs/code drift is common
  • Generating conflict reports for engineering or docs teams prior to publication

Best practices

  • Verify all source URLs/paths and types before running a merge
  • Choose a merge mode that fits risk: rule-based for speed, AI-enhanced for nuance, manual for full control
  • Escalate ambiguous conflicts to a human rather than auto-resolving
  • Persist checkpoints and run merges incrementally on large codebases
  • Filter conflicts by severity to prioritize remediation work

Example use cases

  • Merge v2 docs and v3 code to identify breaking signature changes before release
  • Create a combined API reference for a SDK that pulls docs, README, and inline code comments
  • Audit a repo to surface implemented endpoints that lack documentation
  • Generate a conflict report for a cross-team review meeting
  • Resume a paused merge after a transient source outage using saved checkpoints

FAQ

The skill flags the conflict, marks its severity, and prompts you to choose an authoritative source or supply a resolution strategy; it never silently picks one.

Can the tool run incrementally on very large projects?

Yes. It saves checkpoint files and supports sequential processing, retries, and resume to handle memory limits or partial failures.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational
source-unifier skill by jmagly/aiwg | VeilStrat