brainstorming
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches a branding logo from the external domain 'https://primeradiant.com/brand/superpowers-visual-brainstorming-logo.png' during server initialization.
- [COMMAND_EXECUTION]: The skill executes a local Node.js server using 'scripts/server.cjs' to provide visual mockup and diagram functionality in the user's browser.
- [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection by ingesting interaction data from the visual companion's browser interface.
- Ingestion points: The agent is instructed to read the '$STATE_DIR/events' file to process user selections and clicks from the browser interface.
- Boundary markers: Absent; the skill does not explicitly use delimiters or warnings to ignore instructions embedded within these interaction events.
- Capability inventory: Interaction events are primarily used to influence the dialogue and design process; they do not directly trigger high-risk capabilities like shell execution or file system writes within the brainstorming logic itself.
- Sanitization: The local server validates that incoming WebSocket messages are valid JSON before appending them to the events log.
Audit Metadata