ralph-tui-create-beads

Pass

Audited by Gen Agent Trust Hub on Mar 9, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection through its PRD processing logic. * Ingestion points: Processes user-supplied PRDs in markdown or text format (SKILL.md). * Boundary markers: No delimiters or 'ignore' instructions are present to protect against malicious instructions in the PRD. * Capability inventory: Generates shell commands via the bd CLI and prepares quality gate commands (e.g., pnpm lint) for execution. * Sanitization: None; the skill trusts and extracts commands directly from the source text.
  • [COMMAND_EXECUTION]: The skill generates shell commands for task creation and relies on the execution of developer commands extracted from potentially untrusted PRDs.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 9, 2026, 10:11 PM