CtrlK
BlogDocsLog inGet started
Tessl Logo

systematic-debugging

遇到任何 bug、测试失败或异常行为时使用,在提出修复方案之前执行

64

Quality

78%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/systematic-debugging/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

85%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is a strong, actionable four-phase debugging workflow with explicit validation checkpoints and well-signaled shallow references. Its main weakness is moderate verbosity from restating obvious debugging basics that Claude already knows.

Suggestions

Trim obvious sub-bullets under 阅读错误信息 (e.g. "不要跳过错误或警告", "完整阅读堆栈跟踪") since Claude already knows to read error messages fully.

Consider moving the full multi-component instrumentation bash example into a referenced file and keeping only the decision pattern inline, to reduce token weight in the overview.

Tighten the 常见借口 table or the 常见情况 lists where the same point is restated across sections.

DimensionReasoningScore

Conciseness

Mostly efficient process discipline (red lines, thresholds, excuses table) that Claude would not reliably self-impose, but padded with some obvious sub-bullets like "不要跳过错误或警告" and "完整阅读堆栈跟踪" that explain basics Claude already knows. Not 1 because it is not explaining what a library/format is; not 3 because several passages could be tightened.

2 / 3

Actionability

Provides concrete, executable guidance: numbered steps, an executable multi-layer bash instrumentation example (echo, env, security list-keychains, codesign --verbose=4), and specific numeric thresholds (3 次或以上 → 质疑架构). Not 2 because the code block is real and copy-paste adaptable rather than pseudocode, and the rules are concrete.

3 / 3

Workflow Clarity

Four clearly sequenced phases with an explicit gate ("你必须完成每个阶段后才能进入下一个"), feedback loops (生效→下一阶段;没生效→新假设), verification checkpoints, and a quick-reference checklist table. Not 2 because validation is explicit and woven throughout rather than only listed with gaps.

3 / 3

Progressive Disclosure

SKILL.md acts as a clear overview with well-signaled, one-level-deep references under 辅助技术 (root-cause-tracing.md, defense-in-depth.md, condition-based-waiting.md), each given a one-line description. Not 2 because references are clearly signaled and shallow rather than nested or vague; the body length is well-sectioned.

3 / 3

Total

11

/

12

Passed

Description

72%

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 provides strong, natural trigger terms and a clearly distinct niche for debugging, but is weakened by an implied rather than stated 'what' and limited enumeration of concrete actions. Adding a brief statement of what the skill produces would lift completeness.

Suggestions

State explicitly what the skill does, e.g. add a clause like '系统化定位根本原因并通过四阶段流程修复' so the 'what' is not only implied.

Expand trigger term coverage with common variations a user might say, such as 报错 (error)、崩溃 (crash)、异常 (exception)、堆栈跟踪 (stack trace).

Consider listing 2-3 concrete actions (e.g. 复现问题、定位根因、验证修复) to move specificity toward a fuller enumeration.

DimensionReasoningScore

Specificity

Names the debugging domain and a concrete gating action ("在提出修复方案之前执行") but does not enumerate multiple specific concrete actions the skill performs, matching 'Names domain and some actions, but not comprehensive'. Not 1 because it is concrete rather than vague like 'Helps with documents'; not 3 because there is no list of multiple distinct actions.

2 / 3

Completeness

It has an explicit 'when' clause ("遇到任何 bug...时使用") but the 'what' is only implied — it states a constraint (execute before proposing a fix) without describing what the skill actually produces. Not 1 because the trigger guidance is explicit rather than missing; not 3 because it does not clearly answer what the skill does.

2 / 3

Trigger Term Quality

Terms like "bug"、"测试失败"、"异常行为" are exactly what a user would naturally say when reporting a problem, giving good coverage of natural trigger language. Not 2 because these are the dominant natural phrasings rather than only 'some relevant keywords'; the missing variants (errors, crashes, stack traces) are secondary.

3 / 3

Distinctiveness Conflict Risk

Debugging is a clear, well-bounded niche with distinct triggers (bugs, test failures, abnormal behavior) and a distinctive constraint, making it unlikely to fire for the wrong skill. Not 2 because the triggers are specific and clearly distinct rather than overlapping with many skills.

3 / 3

Total

10

/

12

Passed

Validation

81%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation13 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

metadata_field

'metadata' should map string keys to string values

Warning

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

13

/

16

Passed

Repository
jnMetaCode/superpowers-zh
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.