rc-diagnose

Warn

Audited by Gen Agent Trust Hub on Jun 20, 2026

Risk Level: MEDIUMPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted external data such as log dumps, HAR files, and captured traces, which presents an indirect prompt injection surface where embedded instructions could influence agent behavior.\n
  • Ingestion points: SKILL.md (Phase 1) describes reading HAR files, log dumps, and core dumps.\n
  • Boundary markers: There are no instructions for using delimiters or boundary markers when reading this external data.\n
  • Capability inventory: The agent has capabilities to execute shell commands, run tests, and perform network requests via curl.\n
  • Sanitization: No sanitization or filtering of external debugging artifacts is required or suggested.\n- [COMMAND_EXECUTION]: The skill instructs the agent to dynamically generate and execute executable content as part of the debugging process.\n
  • Script generation: SKILL.md (Phase 1 and Phase 5) directs the creation and execution of throwaway harnesses, bisection scripts, and regression tests.\n
  • Subprocess calls: The skill involves running command-line tools like git bisect, curl, and various test runners based on agent-generated code.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jun 20, 2026, 04:57 AM
Security Audit — agent-trust-hub — rc-diagnose