nested-TAD-detection

Pass

Audited by Gen Agent Trust Hub on Apr 15, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface because it ingests and processes external Hi-C data files. Ingestion points: External file paths and URIs provided via mcool_uri and input_hic. Boundary markers: None; the skill does not specify the use of delimiters or instructions to ignore embedded content within the processed data files. Capability inventory: The skill includes tools for file format conversion, matrix extraction, and TAD calling (mcp__HiCExplorer-tools__hic_to_mcool, mcp__cooler-tools__dump_dense_matrix, mcp__OnTAD-tools__run_ontad). Sanitization: The skill instructions do not specify explicit sanitization or validation logic, relying on the internal implementation of the referenced MCP tools.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 15, 2026, 06:12 AM
Security Audit — agent-trust-hub — nested-TAD-detection