specalign
Align spec files with implementation. Detects drift between spec and code, surfaces discrepancies, user decides whether to update spec or code. Use when both a spec file and its implementation are in context.
90
87%
Does it follow best practices?
Impact
92%
1.12xAverage score across 3 eval scenarios
Passed
No known issues
Quality
Discovery
75%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is a solid description that clearly communicates its purpose and includes an explicit 'Use when' clause with a specific trigger condition. Its main weaknesses are moderate specificity in the actions it performs and limited coverage of natural language variations users might use when needing this skill.
Suggestions
Add more concrete actions, e.g., 'compares function signatures, data models, and API contracts between spec and implementation'
Expand trigger terms to include natural variations like 'specification', 'out of sync', 'spec outdated', 'code doesn't match spec', 'sync spec with code'
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain (spec-code alignment) and some actions (detects drift, surfaces discrepancies), but doesn't list multiple concrete actions in detail—e.g., what kinds of drift, what specific operations it performs on specs or code. | 2 / 3 |
Completeness | Clearly answers both what ('Detects drift between spec and code, surfaces discrepancies, user decides whether to update spec or code') and when ('Use when both a spec file and its implementation are in context') with an explicit trigger clause. | 3 / 3 |
Trigger Term Quality | Includes relevant terms like 'spec file', 'implementation', 'drift', and 'discrepancies', but misses common natural variations users might say such as 'specification', 'out of sync', 'spec outdated', 'code doesn't match spec', or file extensions. | 2 / 3 |
Distinctiveness Conflict Risk | The niche of spec-implementation alignment is quite distinct and unlikely to conflict with other skills. The trigger condition ('both a spec file and its implementation are in context') is specific and narrow. | 3 / 3 |
Total | 10 / 12 Passed |
Implementation
100%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a high-quality skill that clearly defines both passive awareness behavior and an explicit alignment workflow. The drift categories, mutation policy, and numbered discrepancy format provide concrete, actionable structure. The content is concise, well-organized, and includes appropriate validation checkpoints throughout the workflow.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The content is lean and well-structured. Every section earns its place—drift categories are listed concisely as definitions, the mutation policy is tight, and there's no explanation of concepts Claude already knows. No padding or unnecessary context. | 3 / 3 |
Actionability | The skill provides concrete, specific guidance: exact file patterns to search for, a mapping table template, a numbered discrepancy format with stable IDs, and explicit actions for each resolution path (run lint/typecheck, run tests, outline changes before editing). While there's no executable code, this is an instruction-only skill where the guidance is fully actionable. | 3 / 3 |
Workflow Clarity | The 6-step workflow is clearly sequenced with explicit validation checkpoints: run lint/typecheck after code changes, run unit tests if they exist, confirm before non-trivial edits, and a summary step that tracks resolution status. The feedback loop of presenting discrepancies → user decision → apply changes is well-defined. | 3 / 3 |
Progressive Disclosure | The content is well-organized into two clear sections: 'Principles (Always Active)' for passive behavior and 'Workflow (When Explicitly Aligning)' for the active process. For a standalone skill with no bundle files, the structure is appropriate—sections are logically separated, and the content length doesn't warrant splitting into additional files. | 3 / 3 |
Total | 12 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- Repository
- NeverSight/skills_feed
- Commit
aa009ea
- Reviewed
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.