exp-batch

Pass

Audited by Gen Agent Trust Hub on Sep 23, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests untrusted data from external sources and possesses significant tool capabilities.
  • Ingestion points: The agent is instructed to read external experiment logs, epoch counters, and shared project memory in SKILL.md.
  • Boundary markers: No explicit boundary markers or instructions to ignore potentially malicious embedded content within the logs were observed.
  • Capability inventory: The skill allows for command execution (nohup/tmux/launcher) and writes to shared metric tables (SKILL.md).
  • Sanitization: No sanitization or validation logic for the external log data was identified.
  • [COMMAND_EXECUTION]: The skill provides instructions for the agent to execute system commands to manage long-running background tasks. Evidence: 'Launch in the background per project conventions (nohup/tmux/launcher)' in SKILL.md.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 23, 2026, 01:55 AM
Security Audit — agent-trust-hub — exp-batch