meta-skill-importer

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

• Third-party content exposure detected (high risk: 0.85). The required workflow clones an outsider-authored public Git repository at runtime (git clone $RepoUrl $TempDir), then moves the extracted outsider skill directory (including its SKILL.md/README.md) into the local library where it can be ingested as LLM-readable text—indirect prompt injection risk.

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

• Potentially malicious external URL detected (high risk: 0.90). The import script performs a runtime git clone of the supplied RepoUrl (e.g., https://github.com/google-gemini/gemini-skills) and directly copies remote skill files (including SKILL.md) into the local skills library, which lets fetched remote content control agent prompts/instructions.