ClawHub

ExpertPack CLI

v1.0.0

Run ExpertPack CLI tools for validating, fixing, graphing, and deploying packs. Use when: running ep-validate, ep-doctor, ep-graph-export, ep-strip-frontmatt...

⭐ 0· 42·0 current·0 all-time
byBrian Hearn@brianhearn
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 state the skill runs local ExpertPack CLI tools and the SKILL.md only references the local repo (github.com/brianhearn/ExpertPack) and local filesystem operations; nothing requested or described (no env vars, no binaries, no network fetches) is inconsistent with that purpose.
ℹ
Instruction Scope
Instructions confine actions to local tools in ExpertPack/tools/validator/ and correctly advise dry-runs before applying fixes. The doc also suggests copying a .obsidian template and editing an agent config (openclaw.json) for RAG configuration — these are user-facing configuration changes outside the pack itself and should be performed deliberately (not performed automatically by the agent).
✓
Install Mechanism
No install spec and no code files beyond documentation; this is lowest-risk (instruction-only) — nothing is downloaded or installed by the skill.
✓
Credentials
The skill declares no required environment variables, credentials, or config paths; SKILL.md does not instruct the agent to read unrelated secrets or credentials.
✓
Persistence & Privilege
always is false, autonomous invocation is allowed (the platform default) and appropriate for a user-invocable CLI helper; the skill does not request persistent system-wide privileges or modify other skills' configurations.
Assessment
This skill is a set of instructions for running local ExpertPack Python tools — it appears internally consistent. Before installing or using it: (1) ensure you have a trusted clone of the ExpertPack repo at the path you point the tools to (confirm the source on GitHub); (2) always run the doctor/ fixer tools in dry-run mode and inspect proposed changes before using --apply; (3) be cautious about blindly copying templates or editing your agent's openclaw.json — make backups first; (4) run these scripts in a controlled environment (or on a copy of your pack) if you’re unsure about automated fixes. If you want extra assurance, review the actual Python scripts in ExpertPack/tools/validator/ from the upstream repo before applying fixes.

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

latestvk979tw4zsakvnb3mjtzbz6fdnx84sgzt

License

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

Comments