antibody-sequence-fto
Pass
Audited by Gen Agent Trust Hub on Aug 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided antibody sequences and external patent data retrieved from vendor search tools to generate an HTML report, which creates a surface for indirect prompt injection.
- Ingestion points: The agent ingests amino acid sequences provided by users and patent metadata/claims retrieved via
ls_sequence_search_get_resultsandpatsnap_search(SKILL.md). - Boundary markers: The skill instructions do not specify the use of delimiters or 'ignore' instructions to isolate untrusted search results from the agent's core instructions.
- Capability inventory: The skill has the ability to invoke sequence searches, perform patent database queries, and generate a structured HTML report (SKILL.md, manifest.json).
- Sanitization: There is no documented requirement for the agent to sanitize, escape, or validate the content of retrieved patent claims before including them in the final HTML report output.
Audit Metadata