butler

启动 shiji-kb 史记 Wiki 管家的永续 loop。每轮执行一个原子动作（W1选任务→W2执行→W3/W4评估→记账），完成后立即进入下一轮，无需用户逐轮确认。支持多实例并行（--instance/--focus）。

Quality

76%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Critical

Do not install without reviewing

Fix and improve this skill with Tessl

tessl review fix ./.claude/skills/butler/SKILL.md

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 highly actionable, well-sequenced operational spec with explicit validation checkpoints and clean cross-skill navigation; its main weakness is mild redundancy where the output format and lock operations are described in two places.

Suggestions

Remove the duplicate output-format description: keep it only in the '每轮输出格式' section and reference it from startup-flow step 4.

Collapse the startup-flow step 3 one-liner summary of lock ops into a pointer to the '每轮锁操作规范' section to avoid restating the same sequence.

Tighten the '授权声明' list so each authorization is stated once rather than re-anchored across sections.

Dimension	Reasoning	Score
Conciseness	Mostly lean and operational with no concept explanations Claude already knows, but the per-round output format is restated in both the startup-flow step 4 and the dedicated '每轮输出格式' section, and lock ops are summarized then re-detailed; could be tightened. Not level 3 due to this redundancy.	2 / 3
Actionability	Provides fully executable command templates with exact script paths and flags (e.g., 'python3 wiki/scripts/butler/claim_round.py --instance <实例名>'), exit-code semantics ('# exit 1 = 冲突'), and concrete output examples — copy-paste ready, matching the level-3 anchor.	3 / 3
Workflow Clarity	The startup flow and per-round lock ops are clearly sequenced (3a→3b→3c→3d→3e) with explicit validation checkpoints (duplicate-instance check → stop; check-page conflict → skip page) and a feedback loop, plus enumerated pause conditions; matches the level-3 anchor.	3 / 3
Progressive Disclosure	Acts as an overview with a clearly signaled '相关 Skill' section linking one-level-deep to W0/W1/W2/W10/multi-instance docs and a tool-skill table; no nested 2+ level references, matching the well-signaled one-level-deep anchor.	3 / 3
	Total	11 / 12 Passed

Description

67%

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 specific and distinctive about the butler's perpetual-loop behavior, but it omits an explicit 'when to use' trigger clause and relies on internal W1-W4 jargon over natural user keywords.

Suggestions

Add an explicit 'Use when ...' clause stating when to invoke the butler (e.g., 'Use when the user wants to run the 史记 Wiki 管家 loop or maintain shiji-kb autonomously').

Replace or gloss the W1/W2/W3/W4 internal stage codes with natural terms (e.g., 'select task → execute → evaluate → record') so users can recognize the trigger.

Include common natural variations a user might say ('wiki butler', '史记管家', 'run the loop') to broaden trigger coverage.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions in the round loop ('W1选任务→W2执行→W3/W4评估→记账') plus multi-instance parallelism ('--instance/--focus'), matching the 'lists multiple specific concrete actions' anchor.	3 / 3
Completeness	Clearly states what the skill does (perpetual loop, per-round atomic action, multi-instance) but provides no 'Use when...' trigger guidance, so per the rubric completeness is capped at 2 rather than 3.	2 / 3
Trigger Term Quality	Contains some recognizable terms ('史记 Wiki 管家', '永续 loop') but leans on internal jargon (W1/W2/W3/W4) and lacks common natural variations a user would say; not the level-3 broad coverage of natural terms.	2 / 3
Distinctiveness Conflict Risk	Occupies a very specific niche (shiji-kb 史记 Wiki butler perpetual loop) with distinct triggers, making it unlikely to fire for the wrong skill; matches the 'clear niche' anchor.	3 / 3
	Total	10 / 12 Passed

Validation

93%

Warnings & errors only

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

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
relative_links	Relative link issues: 5 suspicious	Warning

	Total	15 / 16 Passed

Repository: baojie/shiji-kb
Path: .claude/skills/butler/SKILL.md
Commit: 79c272f

Reviewed: 4 days ago

Table of Contents

Discovery Implementation Validation

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.