data-contract-extractor
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data in the form of data PRDs, READMEs, and lineage documentation.
- Ingestion points: The agent reads user-provided specifications and design documents as input for extraction, as defined in the overview and step-by-step extraction sections of SKILL.md.
- Boundary markers: The skill lacks explicit instructions or delimiters for the agent to ignore or isolate instructions embedded within the processed documentation to prevent them from being interpreted as agent commands.
- Capability inventory: The skill itself has no defined tools in its frontmatter and demonstrates no access to system commands, network operations, or file-writing capabilities across its scripts, significantly limiting the risk of a successful injection.
- Sanitization: No sanitization, escaping, or validation of the input text is specified before the content is interpolated into the extraction logic or emitted in the YAML output.
Audit Metadata