wf0-nextstep

Pass

Audited by Gen Agent Trust Hub on Feb 28, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill is designed to invoke other commands (e.g., /wf7-pr, /wf5-implement) automatically using the Skill tool. It bypasses user confirmation to streamline the workflow process.
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection as it processes the 'next' field from 'state.json' to determine its next action. 1. Ingestion points: state.json file (specifically the 'next' and 'work-id' fields). 2. Boundary markers: None present to distinguish data from instructions. 3. Capability inventory: Ability to execute any skill defined in the system via string concatenation (/<next_phase>). 4. Sanitization: The skill does not describe any validation or escaping of the values retrieved from the state file before using them in command strings.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 28, 2026, 11:35 PM