audit-spec
Reviews technical documents for ambiguity, inconsistencies, and architectural violations.
Install
mkdir -p .claude/skills/audit-spec && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/15268" && unzip -o skill.zip -d .claude/skills/audit-spec && rm skill.zipInstalls to .claude/skills/audit-spec
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
Audit specification documents for ambiguity, consistency, and architectural compliance.Key capabilities
- →Scan specification documents for vague language
- →Verify alignment between requirements and technical design
- →Ensure compliance with AGENTS.md rules
- →Flag critical issues requiring immediate fixes
- →Generate a Spec Audit Report
How it works
The skill reviews specification documents by scanning for ambiguous language, checking consistency between stories and design, and verifying compliance with architectural rules.
Inputs & outputs
When to use audit-spec
- →Check for ambiguous language
- →Review requirements consistency
- →Verify architectural compliance
- →Identify project risks
About this skill
Skill: Spec Linter (Specification Audit)
<role_gate>
<required_agent>QualityGuard</required_agent>
<instruction>
Before proceeding with any instructions, you MUST strictly check that your ACTIVE_AGENT_ID matches the required_agent above.
Match Case:
- Proceed normally.
Mismatch Case:
- You MUST read the file
.github/agents/{required_agent}.agent.md. - You MUST ADOPT the persona defined in that file for the duration of this skill.
- Proceed with the skill acting as the {required_agent}.
You are supporting the @QualityGuard. Your goal is to review specification documents (Requirements, Design, Plans) to ensure they are clear, consistent, and compliant with project standards before implementation begins.
📋 Task Initialization
IMMEDIATELY use the #todo tool to register the following tasks:
- Ambiguity Check: Scan for vague language.
- Consistency Check: Verify alignment between strict requirements and technical design.
- Architecture Check: Ensure compliance with
AGENTS.mdrules. - Generate Report: Output the Spec Audit Report.
🎯 Objective
Review the provided specification document(s) to prevent "Garbage In, Garbage Out". Ensure the developer has a crystal-clear, feasible, and compliant plan.
🛡️ Audit Steps
1. Ambiguity Check
- Scan for Vague Words: Look for phrases like:
- "maybe", "probably", "roughly", "about"
- "should be able to" (unless strictly defined)
- "user friendly", "fast" (without metrics)
- "etc.", "and so on"
- Action: Flag these as warnings requiring clarification.
2. Consistency Check
- Story vs. Design: Ensure every User Story has a corresponding technical plan (API endpoint, DB schema change, or UI component).
- Missing Technical Details: If a story requires a new field, is the DB migration specified? If it implies a new page, is the route defined?
3. Architecture Compliance (AGENTS.md)
- Check for Violations:
- Does the design use the Repository Pattern? (Required)
- Are Agents incorrectly assigned runtime roles? (Agents are build-time only).
- Are strict rules from
AGENTS.mdfollowed?
📝 Output Format
You must output a Spec Audit Report.
# Spec Audit Report
## 1. Summary
[Brief summary of the spec quality. Is it ready for implementation?]
## 2. Issues & discrepancies
### 🔴 Critical (Must Fix)
- [ ] **Consistency**: User Story "X" implies a database change, but no schema update is listed in the Design section.
- [ ] **Architecture**: The design proposes a direct database query in the controller, violating the Repository Pattern.
### 🟡 Warnings (Clarification Needed)
- [ ] **Ambiguity**: "The system should be fast" - Define "fast" (e.g., < 200ms).
- [ ] **Ambiguity**: "Handle various errors" - List specific error codes to handle.
## 3. Conclusion
[ ] **APPROVE**: The spec is clear, consistent, and compliant. Proceed to Implementation.
[ ] **REQUEST CHANGES**: Address the critical issues above before proceeding.
When not to use it
- →When the `ACTIVE_AGENT_ID` does not match the `required_agent`
- →When the objective is not to review specification documents
- →When the developer already has a clear, feasible, and compliant plan
Limitations
- →Requires the `ACTIVE_AGENT_ID` to match the `required_agent`
- →Focuses on specific vague words and phrases
- →Checks for violations of the Repository Pattern and Agent runtime roles
How it compares
This skill provides a structured audit process for specification documents, explicitly checking for ambiguity, consistency, and architectural compliance, which is more thorough than a general review.
Compared to similar skills
audit-spec side by side with the closest alternatives in the catalog.
| Skill | Installs | Updated | Safety | Difficulty |
|---|---|---|---|---|
| audit-spec (this skill) | 0 | 6mo | No flags | Intermediate |
| deepwiki-rs | 25 | 9mo | Review | Intermediate |
| agui-dotnet-feature-workflow | 0 | 1mo | Review | Advanced |
| architecture-refactor | 0 | 2mo | No flags | Advanced |
Try saying
Example prompts that trigger this skill in your AI assistant.
You might also like
deepwiki-rs
sopaco
AI-powered Rust documentation generation engine for comprehensive codebase analysis, C4 architecture diagrams, and automated technical documentation. Use when Claude needs to analyze source code, understand software architecture, generate technical specs, or create professional documentation from any programming language.
agui-dotnet-feature-workflow
ag-ui-protocol
>
architecture-refactor
SourceLin
用于执行 Sourcelin 跨端架构重构、目录整理、页面拆分、模块边界治理、规则和文档同步。
specification-architect
adrianpuiu
A rigorous, traceability-first system that generates five interconnected architectural documents (blueprint.md, requirements.md, design.md, tasks.md, and validation.md) with complete requirements-to-implementation traceability. Use this skill when users need to architect systems, create technical specifications, or develop structured project documentation with guaranteed traceability.
codebase-documenter
mhattingpete
Generates comprehensive documentation explaining how a codebase works, including architecture, key components, data flow, and development guidelines. Use when user wants to understand unfamiliar code, create onboarding docs, document architecture, or explain how the system works.
c4-architecture-c4-architecture
sickn33
Generate comprehensive C4 architecture documentation for an existing repository/codebase using a bottom-up analysis approach.