psique-workflow-clinic

Warn

Audited by Gen Agent Trust Hub on Jul 12, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The installation instructions direct users to clone a repository from an untrusted GitHub user account ('fahad-hamid') that does not match the skill's stated author ('Aradotso').
  • [PROMPT_INJECTION]: The integrated AI orchestration service processes clinical data, creating a surface for indirect prompt injection. Ingestion points: Clinical notes and patient session history are ingested into LLM prompts in the summarizeClinicalNotes and predictTherapeuticOutcome functions. Boundary markers: The prompts do not use delimiters or boundary instructions to isolate untrusted clinical data from system instructions. Capability inventory: The skill has the capability to trigger WhatsApp messages, generate official AFIP invoices, and manage video session tokens. Sanitization: No input validation or escaping mechanisms are implemented for the clinical data before it is interpolated into LLM prompts.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jul 12, 2026, 03:27 PM
Security Audit — agent-trust-hub — psique-workflow-clinic