metabase-embedding-sso-implementation
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from a remote source which could influence the agent's behavior.
- Ingestion points: The skill fetches version-specific documentation from https://www.metabase.com/docs/v0.{VERSION}/llms-embedding-full.txt (SKILL.md, Step 1a).
- Boundary markers: The instructions do not define delimiters or specific 'ignore embedded instructions' warnings for the fetched documentation.
- Capability inventory: The agent is granted Write, Edit, and Bash tools to modify the project based on the documentation content.
- Sanitization: There is no evidence of sanitization or validation of the fetched documentation before it is used to plan and apply code changes.
- [EXTERNAL_DOWNLOADS]: Fetches configuration and documentation from Metabase's official website.
- Evidence: Downloads llms-embedding-full.txt from metabase.com (a well-known service related to the skill's purpose).
- [COMMAND_EXECUTION]: The skill instructs the agent to install standard software packages.
- Evidence: May execute npm install, pip install, gem install, etc., to add JWT signing libraries (e.g., jsonwebtoken, PyJWT) as needed by the host application.
Audit Metadata