tidb-change-instruction-critic
Use when implementing a user- or reviewer-prescribed code change (including review comments with suggested fixes or options), especially when the requested edit may be risky, incomplete, ambiguous, or misaligned with TiDB correctness and compatibility constraints.
51
55%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Advisory
Suggest reviewing before use
Optimize this skill with Tessl
npx tessl skill review --optimize ./.agents/skills/tidb-change-instruction-critic/SKILL.mdQuality
Discovery
40%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 description focuses almost entirely on when to use the skill but fails to articulate what the skill actually does. It lacks concrete actions or capabilities, reading more like a trigger condition than a complete skill description. The TiDB-specific context provides some distinctiveness, but the absence of specific actions significantly weakens its utility for skill selection.
Suggestions
Add explicit capability statements describing what the skill does, e.g., 'Validates proposed code changes against TiDB correctness constraints, identifies risks in suggested edits, and recommends safe implementations.'
Include more natural trigger terms users would say, such as 'PR review', 'apply suggestion', 'code review feedback', 'patch', or 'fix recommendation'.
Restructure to lead with concrete actions (what it does) before the 'Use when...' clause to clearly separate capabilities from trigger conditions.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description does not list any concrete actions. Phrases like 'implementing a user- or reviewer-prescribed code change' are vague and do not specify what the skill actually does (e.g., validate changes, suggest fixes, rewrite code). It describes a situation rather than concrete capabilities. | 1 / 3 |
Completeness | The description has a clear 'Use when...' clause addressing when to use the skill, but the 'what does this do' part is essentially missing. It describes the triggering context well but never explains what actions the skill performs or what output it produces. | 2 / 3 |
Trigger Term Quality | Includes some relevant terms like 'review comments', 'suggested fixes', 'code change', and 'TiDB' which could match user queries. However, it lacks common natural variations users might say (e.g., 'PR feedback', 'apply suggestion', 'code review', 'patch'). | 2 / 3 |
Distinctiveness Conflict Risk | The mention of 'TiDB correctness and compatibility constraints' provides some domain specificity, but 'implementing a code change' and 'review comments' are broad enough to overlap with general code review or code editing skills. | 2 / 3 |
Total | 7 / 12 Passed |
Implementation
70%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-structured process skill with a strong workflow that includes explicit validation gates and a clear decision framework. Its main weakness is the lack of concrete examples—a sample review comment with a worked-through analysis would significantly improve actionability. The content is reasonably concise but could be tightened in places where steps overlap with the question patterns and output checklist.
Suggestions
Add a concrete worked example: show a sample review comment or change request, then walk through the 7-step workflow with specific analysis, demonstrating the expected output format.
Tighten the Question Patterns section by either integrating the patterns directly into step 6 or making them more distinct from the challenge gate criteria to reduce redundancy.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is reasonably efficient but includes some redundancy—e.g., the 'Question Patterns' section restates what's already covered in the challenge gate step, and some workflow steps could be tightened. It doesn't over-explain concepts Claude knows, but it's not maximally lean either. | 2 / 3 |
Actionability | The skill provides a clear structured workflow and concrete question patterns, but it is entirely instruction-based with no executable code, no concrete examples of input/output (e.g., a sample review comment and the expected analysis), and the guidance remains somewhat abstract—'locate concrete evidence' and 'evaluate solution candidates' lack specificity about how to do so. | 2 / 3 |
Workflow Clarity | The 7-step workflow is clearly sequenced with explicit validation checkpoints: step 2 reconstructs the problem before coding, step 5 requires sharing analysis before implementation, step 6 is a challenge gate that blocks implementation under specific conditions, and step 7 only proceeds after alignment. This includes feedback loops and error recovery patterns. | 3 / 3 |
Progressive Disclosure | For a standalone skill with no bundle files, the content is well-organized into clearly labeled sections (Trigger, Principle, Workflow, Question Patterns, Output Checklist) that are appropriately sized. No monolithic walls of text, no unnecessary nesting, and the structure supports easy scanning. | 3 / 3 |
Total | 10 / 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
- pingcap/tidb
- Commit
e70762e
- 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.