walkthrough-script-agent

Pass

Audited by Gen Agent Trust Hub on Apr 22, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it analyzes untrusted content from external sources such as codebase files, screenshots, and video recordings to identify app features.
  • Ingestion points: The agent ingests data from user-provided codebase, screenshots, and video recordings during the analysis phase.
  • Boundary markers: The skill lacks explicit boundary markers or instructions to disregard potential commands embedded within the analyzed data.
  • Capability inventory: The agent has the ability to read files and execute shell commands via the frame extraction utility.
  • Sanitization: No sanitization or validation is performed on the data extracted from the external source materials.
  • [COMMAND_EXECUTION]: The skill uses a shell command to execute a Python script (extract_frames.py) for processing video recordings. This command is executed using the python3 interpreter and takes a user-specified video path as an argument.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 22, 2026, 07:39 PM