icp-identification

Pass

Audited by Gen Agent Trust Hub on Apr 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute a local Python script (tam_builder.py) via subprocess calls to perform market analysis and data processing tasks.
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface (Category 8) because it ingests and processes untrusted data from the open web.
  • Ingestion points: External content retrieved from user-provided company URLs and web research results during Phase 1.
  • Boundary markers: The instructions do not define explicit delimiters or include warnings for the agent to ignore instructions embedded within the researched web content.
  • Capability inventory: The agent can execute shell commands via python3 to run the tam_builder.py script, providing a path for injected instructions to influence system operations.
  • Sanitization: There is no mention of validation, escaping, or filtering of the external web content before it is synthesized into the research brief or used to define the ICP filters.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 13, 2026, 01:19 PM