switchyard-codebase-exploration

Pass

Audited by Gen Agent Trust Hub on Jul 22, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection by ingesting untrusted data from the codebase and providing execution capabilities.
  • Ingestion points: The skill instructs the agent to search and read content from various directories including switchyard/, tests/, examples/, crates/, and docs/ using rg, find, and git diff.
  • Boundary markers: There are no explicit boundary markers or instructions to the agent to disregard instructions found within the processed files.
  • Capability inventory: The skill provides access to shell commands such as git, rg, find, and uv run pytest, allowing for file system interaction and Python code execution.
  • Sanitization: No sanitization, validation, or filtering of the data retrieved from the files is performed before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 22, 2026, 07:48 AM
Security Audit — agent-trust-hub — switchyard-codebase-exploration