prd-create

Pass

Audited by Gen Agent Trust Hub on Aug 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill performs extensive shell command execution using tools like git, gh (GitHub), glab (GitLab), tea (Forgejo), and the uzi automation binary. It uses these to manage local file changes, perform commits and pushes, and interact with remote issue trackers.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes untrusted user data without explicit sanitization.
  • Ingestion points: User-provided feature concepts, problem descriptions, and solution overviews collected in Step 1.
  • Boundary markers: None identified; user input is directly interpolated into PRD Markdown files and shell commands.
  • Capability inventory: Subprocess execution via git commit, gh/glab/tea issue creation, and uzi binary calls across multiple files and scripts.
  • Sanitization: No sanitization or validation logic is present to escape or filter user input before it is used in commit messages or document bodies.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 13, 2026, 05:40 PM
Security Audit — agent-trust-hub — prd-create