folklore-variant-evidence

Pass

Audited by Gen Agent Trust Hub on Oct 24, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill connects to a remote API at api.helena.bio to fetch genomic variant evidence and literature summaries. These downloads are directed to the official service domain associated with the skill author Helena Bioinformatics.\n- [COMMAND_EXECUTION]: The documentation includes a curl command example to demonstrate how to interact with the JSON-RPC API. This command is restricted to standard POST requests for data retrieval and does not involve executing remote scripts or accessing local files.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes external content retrieved from API calls, representing an indirect injection surface. The skill manages this surface with the following: \n * Ingestion points: Variant evidence, literature associations, and publication details retrieved via MCP tools defined in SKILL.md.\n * Boundary markers: Explicit instructions in the 'Enforce the input boundary' section to prevent the inclusion of patient-specific context.\n * Capability inventory: Network operations via MCP or curl to the vendor's API endpoint as described in SKILL.md and references/mcp-contract.md.\n * Sanitization: Instructions to filter out patient names, case identifiers, and clinical records before making API calls.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 24, 2026, 09:46 PM