CtrlK
BlogDocsLog inGet started
Tessl Logo

systematic-optimization

系统化问题解决与优化流程(领域无关):量化基线→全量列问题→根因分类→结构性方案(拒绝临时补丁)→借鉴同类方案→归纳取舍→展示计划确认实施→实施→验证生效→度量闭环。核心原则:规则/流程存在≠被执行,能落到"无法绕过的机制"就不写建议;优化必须可度量。 用于用户要求"优化""改进""复盘""为什么反复出问题""效率低/成本高/质量差/总是复发"等任何领域的系统性改进,或接手反复失败的任务时。 不用于单点小修复(用 minimal-implementation);不用于只读审查不出方案的场景。

72

Quality

88%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

77%Weight 40%Scale 1-5

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

A well-sequenced methodology skill with strong validation checkpoints and feedback loops, plus concrete decision tables and an output-contract template. Its main weaknesses are mild restatement redundancy across the flow, core-principles, and output-contract sections, and a broken `examples/` reference that weakens progressive disclosure.

Suggestions

Remove the restated content in '核心原则' / '输出契约' that duplicates the nine-step flow, or explicitly frame them as a one-line checklist pointing back to the steps to cut token redundancy.

Fix the `examples/` reference: either create the `examples/` directory with at least one worked example, or remove the dangling reference and inline a brief worked case so the method's application is concrete.

Signal the `examples/` reference as a clear one-level-deep markdown link (e.g., 'See [examples/](examples/)') only once the directory exists, so navigation is unambiguous.

DimensionReasoningScore

Conciseness

Body is dense and assumes Claude's competence, but the '核心原则' section (lines 119-126) and the '输出契约' largely restate the nine-step flow, which could be trimmed. Not 3 because there is no concept-padding (no 'X is a common format that…') and the prose is otherwise lean; not 5 because the repeated restating is a removable redundancy.

4 / 5

Actionability

Concrete, executable guidance for a methodology skill — decision tables (root-cause types, constraint layers), explicit judgment questions ('对每个问题问约束存在吗…'), observable artifacts ('文件 hash / 配置生效 / 版本号'), and a copy-paste output-contract template. Not 5 because worked examples covering common cases are deferred to an external `examples/` path rather than shown inline.

4 / 5

Workflow Clarity

Nine steps (第0步–第9步) are clearly sequenced with explicit validation checkpoints (第8步 三件套: 物证→加载→行为观察) and an explicit error-recovery feedback loop (第9步: '若未改进,回到第2步重新找根因') plus a decision gate (第6步).

5 / 5

Progressive Disclosure

The body has clear section headers and logical flow, but the sole external reference — '相关示例见 examples/' — is a bare backtick path (not a clearly signaled link) pointing to a directory that does not exist; no bundle files are present. Not 4 because the missing/buried reference breaks navigation; not 2 because the inline structure is reasonably organized.

3 / 5

Total

16

/

20

Passed

Description

100%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong description: concrete nine-step actions, natural trigger phrases, explicit when/what coverage, and explicit negative boundaries that distinguish it from sibling skills. Third-person procedural voice with no first/second-person phrasing to penalize.

DimensionReasoningScore

Specificity

Lists multiple specific concrete actions comprehensively — '量化基线→全量列问题→根因分类→结构性方案(拒绝临时补丁)→借鉴同类方案→归纳取舍→…→验证生效→度量闭环' — far beyond vague language.

5 / 5

Completeness

Explicitly answers both 'what' (the full flow plus core principles '约束存在≠被执行…优化必须可度量') and 'when' (positive triggers AND negative scope '不用于单点小修复;不用于只读审查不出方案').

5 / 5

Trigger Term Quality

Comprehensive natural trigger phrases users would actually say — '优化/改进/复盘/为什么反复出问题/效率低/成本高/质量差/总是复发' — with synonym coverage and a clear '用于…' clause.

5 / 5

Distinctiveness Conflict Risk

Clear niche (systematic optimization with the unbypassable-mechanism principle) plus explicit disambiguation naming adjacent skill 'minimal-implementation', minimizing conflict risk.

5 / 5

Total

20

/

20

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.

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
ooooooooooooooooooop/agent-tools
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.