Hermit Search Publish
v1.0.0Local semantic search over knowledge base collections powered by Hermit. Use when: searching knowledge base, semantic search, querying documents, managing co...
⭐ 0· 77·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
The skill is an instruction-only wrapper for the Hermit CLI. Declared required binaries (hermit, uv) and macOS restriction align with the SKILL.md which instructs installing and running the hermit tool. Nothing requested is unrelated to local semantic search.
Instruction Scope
Instructions tell the agent to run hermit commands that recursively scan and index directories the user provides (hermit kb add <directory>). This is consistent with the purpose but means the skill will read any files you point it at — avoid adding broad paths (e.g., your home directory) if you have sensitive data. The SKILL.md also advises deploying the skill into ~/.agents/skills which would make it discoverable by the agent (optional).
Install Mechanism
There is no formal install spec in the registry, but SKILL.md directs using `uv tool install git+https://github.com/xxxgqcoder/hermit.git` and `hermit download`. That will fetch code/models from GitHub and model hubs — standard for this use case but a supply-chain risk: review the upstream repo before running installation commands and be aware model downloads (~1GB) occur.
Credentials
The skill requires no credentials or config paths. It mentions `~/.hermit/` as the data directory (overridable via HERMIT_HOME) but does not demand environment secrets or unrelated tokens.
Persistence & Privilege
always is false and the skill is user-invocable. The SKILL.md suggests an optional location (~/.agents/skills/hermit-search/) to make the skill discoverable, which is normal behavior and limited in scope. The skill does not request system-wide privileges or modify other skills.
Assessment
This skill appears coherent with its purpose, but take these precautions before installing or using it: 1) Review the upstream GitHub repo (https://github.com/xxxgqcoder/hermit) before running the `uv tool install` command to reduce supply-chain risk. 2) Be careful which directories you add as collections — Hermit will recursively read and index any files under the directory you provide, so avoid pointing it at broad or sensitive paths (e.g., your entire home). 3) Expect large model downloads and network access during `hermit download` and model retrieval. 4) If you install the hermit binary from an untrusted source, prefer official releases or verify signatures. 5) Consider running the service in an isolated environment (dedicated user account or container) if you plan to index documents that should not be accessible system-wide.Like a lobster shell, security has layers — review code before you run it.
latestvk97e84nq2fk4tk9fzvc5vp7yfh83crm9
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
Runtime requirements
OSmacOS
Binshermit, uv