github-multi-repo

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime workflow uses GitHub CLI to clone repositories and read their package.json/docs content (e.g., gh repo clone ... then cat .../package.json and gh api .../contents/... --jq '.content' | base64 -d), which can include outsider-authored free text from third-party repos/branches into the agent’s LLM context.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill's runtime snippets explicitly clone and read GitHub repositories (e.g., gh repo clone org/$repo → https://github.com/org/ and gh api repos/:owner/:repo/contents/... → https://api.github.com/repos/:owner/:repo/contents/...) and then run commands like npm install / npm test against the checked-out code, which fetches and executes remote code during skill execution.

MEDIUM W021: Hidden or invisible Unicode characters detected (potential obfuscation or prompt injection).

• Hidden Unicode characters detected (1 type(s) found)