adhd-accountability
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOW
Full Analysis
- [Prompt Injection] (SAFE): Instructional content is focused on task management and accountability. No bypass markers or system-override patterns detected.
- [Data Exposure & Exfiltration] (LOW): The skill uses 'Read', 'Grep', and 'Glob' tools. While these provide access to local files, the absence of network-enabled tools prevents data exfiltration.
- [Indirect Prompt Injection] (INFO): The skill processes untrusted input from task descriptions and file contents. 1. Ingestion points: Reads user conversation and local files via 'Read'. 2. Boundary markers: No delimiters or explicit instructions to ignore embedded content are present. 3. Capability inventory: Read-only access to files; no write or network permissions. 4. Sanitization: No sanitization of external content detected. Severity is low because the skill lacks side-effect capabilities.
- [Remote Code Execution] (SAFE): No remote code downloads or dynamic execution patterns were identified.
Audit Metadata