root-cause-tracing
Pass
Audited by Gen Agent Trust Hub on Jul 1, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The find-polluter.sh script represents a surface for indirect prompt injection as it executes code (npm test) on files identified by a user-controllable search pattern. \n
- Ingestion points: Script arguments POLLUTION_CHECK and TEST_PATTERN in find-polluter.sh. \n
- Boundary markers: Absent; the skill does not explicitly warn the agent to verify these inputs. \n
- Capability inventory: Local command execution (npm test, find, ls) in find-polluter.sh across the local environment. \n
- Sanitization: Absent; parameters are quoted in the script but not validated by the agent. \n- [COMMAND_EXECUTION]: The helper script find-polluter.sh executes npm test to identify side effects in the codebase. While standard for development, this involves executing the project's own code or test suite.
Audit Metadata