jpralph-iterate

Pass

Audited by Gen Agent Trust Hub on Feb 19, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [Prompt Injection] (LOW): The skill is vulnerable to Indirect Prompt Injection (Category 8) because it ingests and acts upon instructions from external data files without sanitization.
  • Ingestion points: prd.json (user stories) and progress.txt (codebase patterns).
  • Boundary markers: Absent. The agent is directed to pick the 'highest priority' story directly from the data.
  • Capability inventory: File system write, git operations, and shell command execution (npm, test, etc.).
  • Sanitization: None provided. A malicious user story in prd.json could potentially trick the agent into performing destructive actions.
  • [Command Execution] (LOW): The instructions command the agent to 'Run quality checks... use whatever your project requires'. While this is the intended purpose of the skill, it creates a surface where a compromised repository's configuration (like package.json scripts) could lead to the execution of malicious code. Severity is lowered as this is the primary intended function of a development agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 19, 2026, 04:15 PM