adhd-assistant

Pass

Audited by Gen Agent Trust Hub on Nov 22, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: No malicious patterns or code detected. The skill consists entirely of instructional markdown and metadata that guide an agent's behavior without executing external scripts or accessing sensitive local system files.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user data, including task lists and productivity notes, which creates a potential surface for indirect prompt injection. This is an inherent functional risk for task-management assistants and is handled through clear behavioral guidelines for the agent.\n
  • Ingestion points: Processes user-provided task descriptions, energy level assessments, and brain dump content via the 'OpenClaw' memory system and local text files.\n
  • Boundary markers: The instructions do not define technical delimiters (e.g., XML tags or triple quotes) for user-provided data.\n
  • Capability inventory: The skill instructions specify the ability to read and write task lists and use scheduling tools for reminders.\n
  • Sanitization: No explicit sanitization or validation of user-provided task content is described in the workflow.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 22, 2026, 01:12 AM