fast-io

Pass

Audited by Gen Agent Trust Hub on Mar 12, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill provides a surface for indirect prompt injection by ingesting and analyzing untrusted content from various sources.
  • Ingestion points: Data enters the agent context via file uploads, external URL imports using the web-import action, user-submitted comments, and markdown notes stored in workspaces as documented in SKILL.md.
  • Boundary markers: The documentation does not specify the use of delimiters or provide instructions for the agent to ignore embedded instructions within these data sources when performing RAG or attachments.
  • Capability inventory: The skill possesses significant capabilities including file modification and deletion via the storage tool, organizational ownership transfer via the org tool, and AI-driven document analysis via the ai tool.
  • Sanitization: There is no evidence of sanitization or filtering of external content before it is processed by the AI or used in tool operations within the provided skill files.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 12, 2026, 05:32 AM