writing-fragments

Pass

Audited by Gen Agent Trust Hub on Oct 26, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill requires the agent to interact with the local file system by reading and appending to a path provided by the user. Without explicit path validation or restriction to a sandbox, an agent could be directed to read from or overwrite sensitive files (e.g., configuration files or SSH keys) if a user provides such a path.- [INDIRECT_PROMPT_INJECTION]: The skill is designed to re-read the contents of the target file from disk before every write operation to preserve user edits. This creates a surface for indirect prompt injection, where malicious instructions placed in the file could influence the agent's behavior during subsequent turns.
  • Ingestion points: Target markdown file content and user dialogue (SKILL.md).
  • Boundary markers: No delimiters or safety instructions are used to distinguish file data from agent instructions.
  • Capability inventory: File system read and write operations (SKILL.md).
  • Sanitization: No validation or sanitization of the file content is performed before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 26, 2026, 09:28 AM