sdd-spec-gc

Audit master specs for stale, provisional, or conflicting requirements.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/fearovex/agent-config --skill sdd-spec-gc
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: sdd-spec-gc
Source: https://github.com/fearovex/agent-config/tree/main/skills/sdd-spec-gc
Command: npx skills add https://github.com/fearovex/agent-config --skill sdd-spec-gc

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Audits master specs to identify stale, provisional, or conflicting requirements.

Core Features & Use Cases

  • Dry-run audits that surface PROVISIONAL, ORPHANED_REF, CONTRADICTORY, SUPERSEDED, and DUPLICATE requirements
  • Interactive user confirmation gate before removals
  • Safe application that preserves all non-removed content and updates changelogs
  • Records GC changes in the spec header and ai-context changelog

Quick Start

Run the sdd-spec-gc audit for a domain to review and prune stale requirements.

Frequently Asked Questions about sdd-spec-gc

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

FAQPage Schema
How do I audit stale specs to identify conflicting or provisional requirements?

You can audit stale specs by running a dry-run report that surfaces PROVISIONAL, ORPHANED_REF, CONTRADICTORY, SUPERSEDED, and DUPLICATE requirements domain-by-domain or across all domains.

Can I preview spec removals before applying changes to the master spec?

Yes, you can preview spec removals using a non-destructive dry-run that presents a report and requires interactive human confirmation before any deletions are applied to preserve all non-removed content.

How do I prune obsolete SDD requirements without losing valid spec data?

You prune obsolete SDD requirements safely by reviewing the dry-run audit report, confirming removals interactively, and letting the process preserve all non-removed content while recording changes in the changelog.

What is the best way to track spec changes when cleaning up orphaned references?

The best way to track spec changes during cleanup is by recording removals in the spec header and ai-context changelog, ensuring all garbage collection changes are documented after confirmation.

Does the spec audit work across all domains or is it limited to a single domain scope?

The spec audit operates both domain-by-domain and across all domains, allowing you to scope the dry-run report to a specific domain or scan the entire master spec for stale requirements.

Why should I use a dry-run audit before removing superseded or duplicate specs?

You should use a dry-run audit because it enforces safety by presenting stale requirements for review and requiring human confirmation before removals, preventing accidental loss of valid spec content.