alterlab-uniprot
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill retrieves protein sequence and functional annotation data from the external UniProt database, which could potentially contain adversarial text intended to influence the agent's behavior.
- Ingestion points: Data is fetched from the UniProt REST API (https://rest.uniprot.org) via the script 'scripts/uniprot_client.py' and referenced code examples.
- Boundary markers: There are no explicit instructions or delimiters designed to isolate the external protein data from the system prompt or agent logic.
- Capability inventory: The skill possesses capabilities for file system access, network requests (WebFetch), and Python script execution (Bash).
- Sanitization: The retrieved data is not subjected to explicit sanitization or validation before being processed by the AI agent.
- [EXTERNAL_DOWNLOADS]: The skill performs network operations to fetch data from rest.uniprot.org. UniProt is an established and well-known scientific service, and these operations are essential for the skill's primary function of bioinformatic data retrieval.
Audit Metadata