ada-detailed-design-audit

Audits detailed-design SD entries against SRS implementation requirements for traceability and verifiability.

Updated Jul 23, 2026
One-click install
npx skills add https://github.com/wubing7755/Ada --skill ada-detailed-design-audit-wubing7755
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ada-detailed-design-audit
Source: https://github.com/wubing7755/Ada/tree/main/skills/software-development/ada-detailed-design-audit
Command: npx skills add https://github.com/wubing7755/Ada --skill ada-detailed-design-audit-wubing7755

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve? Detailed design documents drift from their SRS requirements: entries lose traceability, contain unverifiable vague wording, miss boundary and exception cases, and use inconsistent terminology. This Skill audits SD-* design entries against SRS implementation requirements so every requirement is covered exactly once and every design statement is testable. ## Core Features & Use Cases - Five-dimension audit: Checks verifiability, input/output explicitness, boundary and exception coverage, terminology consistency, and HOW-depth balance, then outputs a problem list with full before/after entry text and waits for explicit approval before editing. - Read-only review modes: Runs requirement-coverage review (six-section report with eight checks), horizontal-consistency review (10 standards), and final delivery gate checks (13 rules) without modifying any files. - Feasibility and gap assessment: Evaluates whether developers can build deliverable software from the document, classifying gaps as document-fixable, external dependencies, or items needing technical spikes. - Use Case: Given an SRS with REQ-F implementation requirements and a detailed design document with SD-1 through SD-61 entries, run the audit to verify zero missing or extra requirement references, symmetric failure-cleanup behavior across sibling entries, and consistent module naming, then apply approved fixes in batches verified by the included Python script. ## Quick Start Ask the agent to audit the detailed design document's SD entries against the SRS implementation requirements for traceability, verifiability, and boundary coverage.

Frequently Asked Questions about ada-detailed-design-audit

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

FAQPage Schema
How do I audit a detailed design document against SRS requirements?

Read both the SRS implementation requirements (REQ-F blocks containing a Derivation field) and all SD design entries, then run the five-dimension audit covering verifiability, input/output explicitness, boundary coverage, terminology, and depth. The included verify_design_doc.py script mechanically checks numbering continuity and REQ coverage set equality.

How to check requirement traceability coverage between SRS and design docs?

Extract the set of SRS implementation requirements and the set of requirement sources cited by design entries, then verify both directions: implementation requirements must be a subset of cited sources (zero uncovered), and cited sources must exist in the SRS at the correct requirement layer. The script reports missing and extra references.

Can the audit run without modifying the design document?

Yes, three read-only modes exist: requirement coverage review producing a six-section report with eight checks, horizontal consistency review with ten standards, and a final delivery check with thirteen rules. All output evidence-backed findings with file and line references without editing files.

Why does requirement extraction report false missing coverage?

Common causes include grep extracting parent requirement IDs from Source fields, requirement layer misclassification by number range instead of the Derivation field, and regex capture of chapter headings into source fields. The documented pitfalls recommend block-based parsing and cross-validation against the requirement analysis checklist.

Does the verification script work with Chinese file paths on Windows?

Yes, the Python script reads files with open(path, encoding='utf-8'), which handles CJK paths correctly. It was written specifically because ripgrep-based search fails on Windows/MSYS with absolute paths containing Chinese characters.