skills/mdsmithaustin/pstack/bro/Gen Agent Trust Hub

bro

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFEPROMPT_INJECTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill contains instructions that direct the agent to "Restate your last message." This pattern is a known technique used in prompt injection attacks to leak system instructions or bypass safety guidelines by forcing the agent to repeat and re-process context that may contain malicious payloads.
  • [INDIRECT_PROMPT_INJECTION]: The skill operates directly on untrusted data from the conversation history without adequate safety boundaries.
  • Ingestion points: The skill ingests the "last message" from the conversation history (SKILL.md).
  • Boundary markers: There are no delimiters or instructions to ignore potential commands embedded within the message being restated.
  • Capability inventory: The skill has the capability to output text, which can be used to reflect malicious instructions back to the agent or the user.
  • Sanitization: No sanitization or validation of the input message is performed before restating it.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 06:52 PM
Security Audit — agent-trust-hub — bro