tidb-failpoint-test-runner
Use when running TiDB package tests and deciding whether failpoint enable/disable is required before and after the test command.
75
62%
Does it follow best practices?
Impact
96%
0.96xAverage score across 3 eval scenarios
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./.agents/skills/tidb-failpoint-test-runner/SKILL.mdQuality
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.
The description is well-targeted to a specific niche (TiDB failpoint management during testing) and includes an explicit 'Use when' trigger clause, making it strong on completeness and distinctiveness. However, it could be more specific about the concrete actions performed (e.g., does it run the tests, inject failpoints, generate commands?) and could include a few more natural keyword variations to improve discoverability.
Suggestions
Add more specific concrete actions, e.g., 'Determines whether to run failpoint-ctl enable/disable around go test commands for TiDB packages, and constructs the appropriate test invocation.'
Include additional trigger term variations such as 'go test', 'make test', 'failpoint-ctl', or 'TiDB unit tests' to improve keyword coverage.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description names a specific domain (TiDB package tests) and mentions a concrete action (deciding whether failpoint enable/disable is required), but it doesn't list multiple specific concrete actions—it's more about a decision process than enumerating capabilities. | 2 / 3 |
Completeness | The description explicitly answers both 'what' (running TiDB package tests and deciding on failpoint enable/disable) and 'when' (starts with 'Use when running TiDB package tests'), providing a clear trigger clause. | 3 / 3 |
Trigger Term Quality | Includes relevant keywords like 'TiDB', 'package tests', 'failpoint', 'enable/disable', and 'test command', which are terms a user working with TiDB would naturally use. However, it misses common variations like 'go test', 'make test', 'tidb testing', or specific failpoint tool names. | 2 / 3 |
Distinctiveness Conflict Risk | This is a very specific niche—TiDB package testing with failpoint management. It is highly unlikely to conflict with other skills due to the narrow domain of TiDB failpoints. | 3 / 3 |
Total | 10 / 12 Passed |
Implementation
50%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill is extremely concise and well-structured as a routing document, but it sacrifices actionability by delegating nearly all concrete guidance to an external file. Without inline commands, decision criteria, or examples, Claude would need to read the referenced document before it could do anything useful. The skill functions more as a table of contents entry than a self-contained instruction set.
Suggestions
Inline the key decision criteria (e.g., how to determine if a package needs failpoint enablement) so the skill is actionable without reading the external file.
Add at least the two canonical command templates (failpoint-enabled and non-failpoint) directly in the skill so Claude can execute them immediately.
Include a brief example showing the decision + command for a concrete package to make the workflow fully concrete.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Very lean — no unnecessary explanations, no padding. Every line serves a purpose. The note about `-tags=intest,deadlock` not enabling failpoints is a non-obvious fact that earns its place. | 3 / 3 |
Actionability | The skill contains no concrete commands, code, or executable examples. It entirely delegates to an external file (`docs/agents/testing-flow.md`) for the actual commands and decision logic, making it a pointer rather than actionable guidance itself. | 1 / 3 |
Workflow Clarity | Steps are listed in a clear sequence, but the actual decision criteria and commands are deferred to an external document. There are no inline validation checkpoints or feedback loops for error recovery. | 2 / 3 |
Progressive Disclosure | References to `docs/agents/testing-flow.md` are clearly signaled with specific section names, which is good. However, since no bundle files are provided, we cannot verify these references exist, and the skill is almost entirely dependent on the external file — it provides very little standalone value. | 2 / 3 |
Total | 8 / 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.