ClawHub

Knowledge to ClawHub Skill Publisher

v1.0.1

Turn a solved workflow into canonical knowledge docs and a self-contained ClawHub-publishable skill. 把一个已验证流程沉淀成规范知识文档，并打包成可发布到 ClawHub 的自包含 skill。

⭐ 0· 96·0 current·0 all-time
by@grey0758
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 match the actions described in SKILL.md and README: extracting workflow docs, assembling a self-contained skill folder, and publishing to ClawHub. It does not ask for unrelated binaries, services, or credentials.
✓
Instruction Scope
Runtime instructions are narrowly focused on reading repo docs (README.md, WORKFLOW.md, FAQ.md, CHANGELOG.md), creating a skill folder under skills/shared/<skill-name>, and invoking clawhub publish/inspect. The SKILL.md explicitly warns not to publish secrets or raw transcripts. One mildly ambiguous step is 'sync the OpenClaw memory mirror' which is mentioned but not specified; otherwise instructions stay within the stated packaging/publishing scope.
✓
Install Mechanism
No install spec and no code files — instruction-only. No downloads or archive extraction are present, so there is no install-time execution risk.
✓
Credentials
The skill declares no required environment variables or credentials. Note: publishing via the clawhub CLI will use whatever ClawHub authentication the user has configured locally (the skill does not request or embed credentials itself), so users should ensure their local clawhub auth is valid and that they do not inadvertently publish sensitive files.
✓
Persistence & Privilege
always:false and there is no mechanism to force permanent presence or modify other skills' configs. The included agents/openai.yaml sets allow_implicit_invocation:false, reducing implicit autonomous invocation in that agent metadata.
Assessment
This skill is coherent and instruction-only, but review the files it will package before publishing. Ensure README.md, WORKFLOW.md, FAQ.md, and CHANGELOG.md contain no secrets, machine-specific tokens, or raw chat transcripts. Publishing requires the clawhub CLI and uses your local ClawHub credentials — confirm your local auth and inspect the remote package after publishing. Be cautious about the unspecified 'sync the OpenClaw memory mirror' step (confirm what it does in your environment) and run the provided clawhub inspect command to verify files and versions before marking a release as final.

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

bilingualvk97844bc1ve7gvjktwsrbyhran8378sxclawhubvk97844bc1ve7gvjktwsrbyhran8378sxknowledgevk97844bc1ve7gvjktwsrbyhran8378sxlatestvk97844bc1ve7gvjktwsrbyhran8378sxskillvk97844bc1ve7gvjktwsrbyhran8378sxworkflowvk97844bc1ve7gvjktwsrbyhran8378sx

License

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

Comments