ralph-tui-create-json
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data (PRDs) to generate structured configuration files used for command execution.
- Ingestion points: The skill ingests PRD content from markdown files or text input provided by the user (SKILL.md).
- Boundary markers: The instructions lack explicit boundary markers or warnings for the agent to ignore potentially malicious instructions embedded within the PRD text.
- Capability inventory: The skill extracts commands from the "Quality Gates" section (e.g.,
pnpm typecheck,pnpm lint) and places them into theacceptanceCriteriaarray of theprd.jsonoutput. These criteria are designed to be executed by theralph-tuiCLI tool. - Sanitization: There is no evidence of sanitization, validation, or escaping of the extracted command strings. An attacker could craft a PRD with a malicious quality gate command that performs unauthorized actions when processed by the downstream tool.
Audit Metadata