subagent-driven-development

Pass

Audited by Gen Agent Trust Hub on Oct 1, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: No malicious patterns or security vulnerabilities were identified. The skill follows best practices for AI-driven development by using fresh subagent contexts for each task and implementing strict quality gates.- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external implementation plans and task descriptions, which represents a standard indirect prompt injection surface for code-generation agents.
  • Ingestion points: Task descriptions and plan context are ingested into the implementer and reviewer subagent prompts (defined in implementer-prompt.md and spec-reviewer-prompt.md).
  • Boundary markers: Prompt templates use distinct markdown headers (e.g., ## Task Description, ## Context) to separate instructions from the data to be processed.
  • Capability inventory: The implementer subagent is granted capabilities to write code, run tests, and commit changes to the repository.
  • Sanitization: The workflow incorporates two independent review stages (spec-reviewer-prompt.md and code-quality-reviewer-prompt.md) to verify that the implementer only performs the requested tasks and adheres to quality standards.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 1, 2026, 12:55 PM