catalog-babysitter-users

Pass

Audited by Gen Agent Trust Hub on Jul 9, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes gh search code and gh api commands to discover repositories and fetch their metadata. These tools are used as intended for repository discovery and management.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted data from external sources (public GitHub repositories). \n
  • Ingestion points: Untrusted data enters the agent context via gh search results and API responses (SKILL.md). \n
  • Boundary markers: No specific delimiters or boundary markers are used to separate the external data from the skill's instructions in the generated docs/repo-with-babysitter-processes.md file. \n
  • Capability inventory: The skill possesses the capability to write to the file system and execute shell commands via the gh tool. \n
  • Sanitization: There is no evidence of sanitization, filtering, or validation of the repository descriptions or topics before they are written to the catalog file.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 9, 2026, 09:58 AM
Security Audit — agent-trust-hub — catalog-babysitter-users