omh-codebase-onboarding

Pass

Audited by Gen Agent Trust Hub on Sep 25, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill analyzes external repositories which may contain malicious instructions embedded in code or documentation.\n
  • Ingestion points: The skill ingests codebase content from the repository root or supplied source context as defined in the 'Required inputs' section.\n
  • Boundary markers: It explicitly establishes a 'read-only boundary' and a 'not-evidence boundary' to distinguish between raw source data and agent-generated findings.\n
  • Capability inventory: Capabilities are limited to codebase search (grep, ripgrep, ast-grep) and runtime status recording via the omh CLI. The skill strictly forbids file mutations, dependency installations, or script execution during the onboarding phase.\n
  • Sanitization: It mandates redacting secrets from configuration or environment files, recording only redacted paths and risk categories.\n- [COMMAND_EXECUTION]: The skill utilizes the omh CLI tool for status recording and standard search utilities (grep, ripgrep, ast-grep) for codebase analysis. These commands are executed within a restricted, read-only context intended for repository exploration, with explicit instructions not to perform setup or installation tasks.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 25, 2026, 08:33 PM
Security Audit — agent-trust-hub — omh-codebase-onboarding