flowio

Pass

Audited by Gen Agent Trust Hub on Aug 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of the 'flowio' package from the public Python Package Index (PyPI) to handle FCS file parsing.
  • [COMMAND_EXECUTION]: The installation instructions utilize shell commands (e.g., uv pip install flowio) to set up the necessary environment.
  • [PROMPT_INJECTION]: The skill ingests data from FCS files, which contain metadata segments (TEXT and ANALYSIS) that can include arbitrary strings from untrusted sources. This creates a surface for indirect prompt injection if the agent interprets or acts on these metadata values.
  • Ingestion points: FCS file ingestion via the FlowData class defined in SKILL.md and references/api_reference.md.
  • Boundary markers: The skill does not implement delimiters or explicit instructions for the agent to ignore embedded commands within the file metadata.
  • Capability inventory: The skill possesses file-writing capabilities through methods like create_fcs, write_fcs, and to_csv documented across both files.
  • Sanitization: There is no documented sanitization, validation, or schema enforcement for the metadata keywords extracted from the processed files.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 17, 2026, 02:12 AM