brainstorm

Pass

Audited by Gen Agent Trust Hub on Apr 19, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill performs legitimate research-related tasks, including reading workspace notes and writing hypotheses to a designated file. It does not engage in network exfiltration or dangerous command execution.\n- [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface because it processes untrusted data from research papers and literature files.\n
  • Ingestion points: The skill reads various Markdown files and paper summaries from reaper-workspace/notes/ and reaper-workspace/papers/, as well as a user-provided [context-hint] argument.\n
  • Boundary markers: The instructions do not specify any delimiters or safety warnings for the agent when processing this external data.\n
  • Capability inventory: The skill's capabilities are restricted to reading research notes and writing new hypotheses to reaper-workspace/notes/ideas.md. It does not utilize network or subprocess execution tools.\n
  • Sanitization: No explicit sanitization or validation of the input content is performed.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 19, 2026, 10:26 AM
Security Audit — agent-trust-hub — brainstorm