persona-health-butler
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill utilizes
npx @agentplanet/acnto facilitate on-chain identity registration (ERC-8004) on the Base network. This involves downloading and executing code from the npm registry. - [COMMAND_EXECUTION]: The skill executes a local script
scripts/state-sync.jsand uses framework-specific CLI commands (openpersona) to read, write, and signal state changes to the host environment. - [INDIRECT_PROMPT_INJECTION]: The skill acts as an attack surface for indirect prompt injection as it ingests and processes untrusted user data (dietary logs, exercise details, mood notes) to generate holistic health reports.
- Ingestion points: User-provided inputs regarding meals, workouts, and emotional states processed by the diet-tracker, exercise-planner, and mood-journal skills.
- Boundary markers: None explicitly implemented to delimit user-provided health data from system instructions.
- Capability inventory: The skill has access to
WebFetch,curl, and file system write operations. - Sanitization: No explicit sanitization of ingested user data was found in the provided configuration files.
Audit Metadata