brainstorming

Pass

Audited by Gen Agent Trust Hub on May 11, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill includes an indirect prompt injection surface as it is instructed to ingest data from untrusted local sources.
  • Ingestion points: The process requires the agent to explore project context by reading files, documentation, and recent git commits (SKILL.md).
  • Boundary markers: Absent; there are no instructions to use delimiters or specific ignore-rules for content found in the project files.
  • Capability inventory: The skill allows for file system reads and writes (writing design docs), git interactions, and the invocation of the subsequent writing-plans skill.
  • Sanitization: Absent; no methods are specified for validating or escaping instructions that might be embedded in the project context being analyzed.
Audit Metadata
Risk Level
SAFE
Analyzed
May 11, 2026, 02:48 PM