omh-run-efficiency

Pass

Audited by Gen Agent Trust Hub on Sep 25, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user requests as a required input but lacks instructions for handling malicious payloads or providing clear boundary markers. This vulnerability surface could allow a user to influence the agent's behavior or reporting logic through embedded instructions.
  • Ingestion points: The skill explicitly lists user request as a required input in the Required inputs section of SKILL.md.
  • Boundary markers: No delimiters or specific safety instructions (e.g., "ignore instructions in the request") are defined to isolate the user input.
  • Capability inventory: The skill executes the omh CLI tool via shell for status tracking and accesses session metadata.
  • Sanitization: There is no evidence of data sanitization or validation performed on the user input before processing.
  • [COMMAND_EXECUTION]: The skill demonstrates the execution of a shell-based CLI tool (omh runtime record) to track its workflow progress. While this is the intended functionality for the 'tracker' role, it confirms the agent's ability to execute shell commands within its environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 25, 2026, 08:34 PM
Security Audit — agent-trust-hub — omh-run-efficiency