writing-fragments

Pass

Audited by Gen Agent Trust Hub on Sep 11, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input from conversations to generate content, representing a standard injection surface.
  • Ingestion points: User conversation text is mined for raw fragments.
  • Boundary markers: The instructions do not specify delimiters to isolate user-provided data from agent instructions.
  • Capability inventory: The skill utilizes basic file system read and write operations for markdown management; model invocation is explicitly disabled via the frontmatter.
  • Sanitization: No explicit sanitization or escaping of captured text is implemented.
  • [SAFE]: No malicious patterns such as command execution, network exfiltration, or persistence mechanisms were detected in the instructions. The skill operates within a limited scope to perform text processing and file storage.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 11, 2026, 02:00 AM
Security Audit — agent-trust-hub — writing-fragments