plan-report

Pass

Audited by Gen Agent Trust Hub on Jul 28, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface by collecting untrusted user input describing project backgrounds and constraints, which is then interpolated into structured report templates. * Ingestion points: User-provided project information in Stage 1. * Capability inventory: The agent is instructed to write files to the 'docs/plan/' directory. * Boundary markers: The instructions lack explicit markers or warnings to disregard potential commands embedded in the provided project data. * Sanitization: There is no mention of sanitizing or validating user input before it is used to generate documentation.
  • [COMMAND_EXECUTION]: The skill provides 'tools/md2html.py', a script for document processing. This utility performs file system read and write operations on paths provided via command-line arguments.
  • [EXTERNAL_DOWNLOADS]: The skill documentation advises the user to install the 'markdown' package from the official Python Package Index to facilitate report conversion.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 28, 2026, 07:06 AM
Security Audit — agent-trust-hub — plan-report