writing-fragments

Pass

Audited by Gen Agent Trust Hub on Aug 27, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill defines a process where the agent reads the current state of a local markdown file before appending new fragments, creating a surface for ingesting untrusted data.\n
  • Ingestion points: Content is read from a user-specified markdown file path on the local filesystem as defined in SKILL.md.\n
  • Boundary markers: Absent. There are no instructions to use delimiters or ignore instructions found within the ingested file content.\n
  • Capability inventory: The skill utilizes both read and write file system access to manage writing fragments.\n
  • Sanitization: Absent. The agent is explicitly told to preserve the file's existing content and modifications, which prevents filtering of potentially malicious instructions embedded in the document.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 27, 2026, 07:25 AM
Security Audit — agent-trust-hub — writing-fragments