senpi-account-status
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data from the Senpi MCP server, which constitutes a potential attack surface for indirect prompt injection if the remote data were compromised.
- Ingestion points: The
status.pyscript retrieves user-specific data including points, rank, and referral balances fromhttps://mcp.prod.senpi.ai/mcp. - Boundary markers: The
SKILL.mdfile provides clear instructions on how to format the output card, though it does not specify explicit delimiters for the data values themselves. - Capability inventory: The skill uses a local Python engine (
status.py) to perform network requests and provide JSON data for the agent to narrate. - Sanitization: The script performs type validation (converting values to floats) and filters for specific fields, which reduces the risk of malicious payload execution, though it does not explicitly sanitize strings against instruction injection.
Audit Metadata