alterlab-opentargets
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from the external Open Targets Platform GraphQL API, which presents a surface for potential indirect prompt injection if the source data were to contain malicious instructions aimed at influencing the agent's behavior.
- Ingestion points: Data is retrieved from the api.platform.opentargets.org endpoint using the execute_query function within the scripts/query_opentargets.py file.
- Boundary markers: The skill instructions do not explicitly define boundary markers or provide specific guidance to the agent on how to treat the retrieved text as data rather than instructions.
- Capability inventory: The skill possesses network access via the requests library and the ability to execute code via the Bash(python:*) tool.
- Sanitization: The technical implementation returns raw dictionary and list structures from the API response without specific sanitization of the textual content (e.g., gene descriptions, literature snippets, or clinical trial summaries) for potential prompt injection content.
Audit Metadata