Security audit
Security checks for vulnerabilities and agentic risk
This school-search skill does what it says, but users should understand it sends location searches to Camino's API and its install examples are not pinned.
Install only the specific school-finder skill from a version or commit you trust, avoid pasting highly sensitive exact home or child-related locations unless necessary, and keep the Camino API key scoped and revocable. Treat the script as a remote API client, not a fully local school-search tool.
SKILL.md:13Unpinned Third-Party Installation Commands Create Supply-Chain Risk
scripts/school-finder.sh:34Unvalidated Parameters Permit URL Query Manipulation and Curl Glob Expansion
Skill reads from agent configuration directories (.claude/, .codex/, .gemini/). These directories may contain API keys, personal settings, and other credentials that the skill has no legitimate need to access.
Add your key to Claude Code:
Add to your ~/.claude/settings.json:
{
The skill documents shell-based execution paths using curl/jq and script invocation, but it does not declare an explicit tool scope or permissions boundary. That increases the chance an agent will run shell commands with broader-than-necessary access, making misuse or unintended command execution harder to constrain.
Using npx skills add without a pinned version makes installation non-reproducible and exposes users to supply-chain risk if the upstream package changes or is compromised. A future package version could execute unexpected code during install or alter skill behavior silently.
This installation command also references npx skills without version pinning, so users may fetch whatever the latest package resolves to at install time. That creates a preventable supply-chain exposure during skill installation.
npx clawhub@latest explicitly tracks the newest published version, which maximizes exposure to malicious or breaking upstream changes. For an installation path, this is a classic supply-chain hardening gap.
The skill is specifically built to send user-provided addresses, queries, and coordinates to a remote Camino API, yet it does not prominently warn users that sensitive location data leaves the local environment. Because home, school, and commute searches can reveal highly sensitive routines and minors-related context, the absence of a clear privacy notice increases privacy and compliance risk.
The skill explicitly sends queries and precise coordinates to https://api.getcamino.ai/, which is an external transmission of potentially sensitive location data. This is expected for the skill's function, but it still creates privacy risk if users are unaware or if highly sensitive addresses are submitted.
curl -H "X-API-Key: $CAMINO_API_KEY" \
"https://api.getcamino.ai/query?query=schools&lat=40.7589&lon=-73.9851&radius=2000&rank=true"
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
curl -s -X GET \
-H "X-API-Key: $CAMINO_API_KEY" \
-H "X-Client: claude-code-skill" \
"https://api.getcamino.ai/query?${QUERY_STRING}" | jq .
No suspicious patterns detected.