Expertise in detecting inconsistencies, gaps, and conflicts across specification documents. Activates when user asks about document quality or consistency. Trigger keywords: consistency, analysis, gap analysis, conflict detection, spec review, cross-reference, traceability
Analyzes specification documents for gaps, conflicts, and traceability issues to ensure alignment.
/plugin marketplace add datamaker-kr/synapse-claude-marketplace/plugin install speckit-helper@synapse-marketplaceThis skill is limited to using the following tools:
references/analysis-rules.mdreferences/severity-levels.mdThis skill provides expertise in detecting inconsistencies, gaps, and conflicts across specification documents. It performs read-only cross-document analysis to ensure that requirements, tasks, data models, and plans are aligned and complete. The output is a structured analysis report with categorized findings and severity levels.
IMPORTANT: This skill NEVER modifies files. It is strictly a read-only analysis tool. All findings are reported for human review and resolution.
The skill is triggered when the conversation involves:
The skill evaluates documents against eight rule categories:
Verify that every functional requirement (FR-XXX) and non-functional requirement (NFR-XXX) in the specification is addressed by at least one implementation task or plan item. Flag orphaned requirements with no downstream mapping.
Ensure every task in tasks.md traces back to a specific requirement or user story in spec.md. Flag tasks with missing [Spec §X.Y] references or invalid reference targets.
Check that the implementation plan is consistent with the task list and specification. Detect mismatches in scope, ordering assumptions, or phase assignments between documents.
Verify that entity definitions, field names, types, and relationships are consistent across the specification, data model section, and API contracts. Flag naming mismatches, type conflicts, or missing fields.
Ensure every API endpoint defined in the specification has corresponding tasks for implementation and testing. Flag endpoints missing from the task list or with incomplete request/response schema definitions.
If a project constitution exists, verify that the specification and tasks comply with its architectural principles, technology constraints, and coding conventions.
Identify duplicate or near-duplicate requirements, tasks, or definitions across documents. Flag redundancies that may lead to conflicting implementations or wasted effort.
Scan for vague, unmeasurable, or subjective language in requirements (e.g., "fast," "user-friendly," "scalable"). Flag ambiguous terms that need quantification or clarification.
Each finding is assigned a severity level:
| Severity | Meaning | Action Required |
|---|---|---|
| CRITICAL | Blocking issue that prevents correct implementation | Must resolve before proceeding |
| HIGH | Significant gap or conflict likely to cause defects | Should resolve before implementation |
| MEDIUM | Inconsistency that may cause confusion or rework | Resolve during implementation |
| LOW | Minor style or convention issue | Resolve at convenience |
The analysis report is structured as follows:
# Consistency Analysis Report
## Summary
- Total findings: N
- Critical: N | High: N | Medium: N | Low: N
## Findings
### [CRITICAL] RC-001: FR-012 has no implementing task
- **Category**: Requirement Coverage
- **Location**: spec.md FR-012
- **Details**: Payment retry logic requirement has no corresponding task in tasks.md.
- **Recommendation**: Add a task in the Stories phase covering FR-012.
### [HIGH] TT-001: Task T008 references non-existent FR-099
- **Category**: Task Traceability
- **Location**: tasks.md T008
- **Details**: The [Spec FR-099] reference does not match any requirement.
- **Recommendation**: Correct the reference or add the missing requirement.
Each finding includes an ID, category, location, detailed description, and a recommended resolution.
For detailed analysis rules, severity definitions, and configuration options, consult:
references/analysis-rules.md -- Full rule definitions with detection logic for each categoryreferences/severity-levels.md -- Severity classification criteria and escalation guidelinesActivates when the user asks about AI prompts, needs prompt templates, wants to search for prompts, or mentions prompts.chat. Use for discovering, retrieving, and improving prompts.
Search, retrieve, and install Agent Skills from the prompts.chat registry using MCP tools. Use when the user asks to find skills, browse skill catalogs, install a skill for Claude, or extend Claude's capabilities with reusable AI agent components.
Creating algorithmic art using p5.js with seeded randomness and interactive parameter exploration. Use this when users request creating art using code, generative art, algorithmic art, flow fields, or particle systems. Create original algorithmic art rather than copying existing artists' work to avoid copyright violations.