ClawHub

experience-model-builder

v1.0.0

将用户提供的经验或方法提炼为结构化、可复用的经验模型，包含角色定位、分析模块、指令约束和调用格式。

⭐ 0· 31·0 current·0 all-time
by@96441113
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
✓
Purpose & Capability
Name/description (提炼经验为结构化模型) align with SKILL.md: the instructions only ask for user-provided practice descriptions and metadata to build a model. There are no unrelated requirements (no cloud keys, no system access).
✓
Instruction Scope
Runtime instructions are limited to collecting user inputs, validating completeness, building a structured model, and outputting Markdown. They do not reference reading files, environment variables, external endpoints, or other system state outside the user-provided content. Rejection rule for 'unhealthy' content is explicit.
✓
Install Mechanism
No install spec and no code files — instruction-only skill. This minimizes disk writes and external code execution risk. The assets/examples/references/scripts folders are placeholders with no executable content.
✓
Credentials
Skill requires no environment variables, credentials, or config paths. The information it requests from users (practice description, optional references, data/cases) is proportionate to the stated goal.
✓
Persistence & Privilege
always:false and default autonomous invocation settings are present; this is normal. The skill does not request persistent system presence, nor does it modify other skills or agent-wide config.
Assessment
This skill is instruction-only and appears coherent with its description — it only needs the user's descriptions and optional supporting info. Before installing: (1) Be cautious about sharing proprietary or sensitive internal processes when giving input (the skill asks you to paste full practices). (2) Review generated models for accuracy and for any inadvertent sensitive detail before sharing externally. (3) Note the skill may ask up to 3 clarifying questions — avoid providing secrets in responses. (4) Because it produces structured, reusable templates, test it first with non-sensitive example data to ensure output format and assumptions meet your needs.

Like a lobster shell, security has layers — review code before you run it.

latestvk976wdv9bx3ynp8gcp3zsc9xk1846w7d

License

MIT-0
Free to use, modify, and redistribute. No attribution required.
Termshttps://spdx.org/licenses/MIT-0.html

Comments