Location Context

Skill name/description match behavior: the script and docs call api.getcamino.ai/context and return location context. The only mismatch is documentation/metadata: the runtime script requires curl and jq but the registry metadata lists no required binaries; otherwise credentials and endpoints align with purpose.

SKILL.md and scripts limit actions to: obtaining/using CAMINO_API_KEY, calling Camino API endpoints (trial and /context), and printing JSON. They instruct adding the key to ~/.claude/settings.json (a config path not declared in metadata) and show npx install steps; they do not request unrelated files or other secrets. This is scoped to the stated purpose, but the documentation should declare the config path and required binaries.

There is no formal install spec inside the skill bundle (instruction-only), but SKILL.md suggests using npx to pull a GitHub repo or clawhub to install. Fetching code from a GitHub repo via npx is common but carries the usual risks of executing remote code; the included script itself is small and benign.

Only CAMINO_API_KEY is required and used. The script uses that key solely to authenticate requests to api.getcamino.ai. No unrelated credentials or broad secrets are requested.

always is false and the skill is user-invocable; it does not request persistent/privileged system changes. It suggests adding the API key to Claude Code settings (user config) which is a normal mechanism for storing credentials for a skill.