CtrlK
BlogDocsLog inGet started
Tessl Logo

crew-creator

Manage and optimize custom agent definition files (IDENTITY.md, AGENTS.md, SOUL.md, TOOLS.md). Also handles first-time employee initialization when .workspace/ has no IDENTITY.md. Use when users want to edit agent identity, modify workflow instructions, adjust personality, add/remove tools, optimize prompts, or initialize a new employee from scratch. Trigger signals: 'modify prompt', 'change identity', 'add tool', 'remove tool', 'optimize workflow', 'adjust personality', 'initialize employee', '修改提示词', '改身份', '加工具', '去掉工具', '优化能力', '调性格', '初始化员工', '创建员工'. Do NOT use for: skill creation (use skill-creator), skill searching (use find_skills tool).

68

Quality

83%

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

SKILL.md
Quality
Evals
Security

Quality

Content

72%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.

Highly actionable and well-structured content with strong progressive disclosure to real bundle files, but it loses points on conciseness due to repeated multilingual-mode explanations and on workflow clarity because the editing flow lacks an explicit validation checkpoint before overwriting agent definition files.

Suggestions

Consolidate the single-language vs multilingual-mode rules into one section; remove the duplicate explanations in 'Generating Files' and 'Multilingual Content Format' that restate the 'Language Generation Strategy' guidance.

Add an explicit validation checkpoint to the Editing Workflow before the write step (e.g., verify the edited content against references/crew-file-format.md and confirm required YAML fields are present) to lift workflow clarity above the destructive-operation cap.

Trim ambient framing around code blocks (e.g., 'The script generates files based on the config') where the code and table already convey the action.

DimensionReasoningScore

Conciseness

Mostly efficient use of tables and code blocks, but the single-vs-multilingual-mode logic is restated in three separate sections ('Language Generation Strategy', 'Generating Files', 'Multilingual Content Format') and the language-mode instructions contain some redundant explanation Claude could infer, fitting the level-3 'mostly efficient but could be tightened' anchor.

3 / 5

Actionability

Provides copy-paste-ready, executable code blocks (write_file + shell_exec init calls, exact tools.py list/detail/search commands) with a concrete JSON config example and a clear file-responsibility table, covering the common cases per the level-5 anchor.

5 / 5

Workflow Clarity

The init flow is clearly sequenced (detection -> information gathering rounds -> generate -> read-back -> adjustments), but the editing workflow overwrites definition files without an explicit validate/verify checkpoint before writing; per the destructive/batch-operations cap, missing validation holds this at level 3 even though sequencing is otherwise strong.

3 / 5

Progressive Disclosure

The body is a clear overview that signals one-level-deep references to three real reference files (crew-file-format, prompt-engineering-guide, available-tools) and three scripts, each linked from a dedicated Reference Documents section, matching the level-5 anchor for well-signaled, easy navigation.

5 / 5

Total

16

/

20

Passed

Description

95%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, well-structured description that covers what, when, and not-when with concrete trigger phrases in two languages and explicit boundary guidance. The only soft spot is the slightly abstract 'manage and optimize' lead verb, which keeps specificity just below a perfect score.

DimensionReasoningScore

Specificity

Names the four concrete target files and several concrete actions (edit identity, add/remove tools, optimize prompts, initialize), but the leading 'Manage and optimize' framing is slightly more abstract than a fully enumerated action list, leaving minor gaps versus the level-5 anchor.

4 / 5

Completeness

Explicitly answers 'what' (manage/optimize four agent definition files plus first-time initialization) and 'when' with concrete trigger phrases, and adds negative boundary guidance ('Do NOT use for: skill creation, skill searching'), satisfying the level-5 anchor.

5 / 5

Trigger Term Quality

Comprehensive natural trigger phrases ('modify prompt', 'change identity', 'add tool', 'remove tool', 'optimize workflow', 'adjust personality', 'initialize employee') plus Chinese synonyms and variants, matching the level-5 anchor's expectation of synonyms and broad coverage.

5 / 5

Distinctiveness Conflict Risk

The explicit 'Do NOT use for' clause carves out a clear niche (agent definition files) and disambiguates from skill-creator and find_skills, giving minimal conflict risk per the level-5 anchor.

5 / 5

Total

19

/

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
dtyq/magic
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.