using-woostack

Pass

Audited by Gen Agent Trust Hub on Jun 15, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill contains instructions that mandate the agent to treat external, project-specific data as authoritative and binding.
  • Ingestion points: The agent is instructed to read the root AGENTS.md file and follow its "woostack section as binding project policy."
  • Boundary markers: There are no defined delimiters or instructions to ignore embedded malicious prompts within the AGENTS.md file.
  • Capability inventory: While this specific skill is primarily for routing, it directs the agent to load other skills with significant capabilities, such as woostack-execute (code execution), woostack-init (workspace modification), and woostack-build (feature construction).
  • Sanitization: The skill lacks any mechanisms for sanitizing or validating the content of the AGENTS.md file before it is adopted as binding policy.
  • [PROMPT_INJECTION]: The skill contains explicit override markers that prioritize the AGENTS.md file over all other instructions, including "generic agent defaults."
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 15, 2026, 04:06 PM
Security Audit — agent-trust-hub — using-woostack