subagent-driven-development

Pass

Audited by Gen Agent Trust Hub on Apr 12, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection vulnerability surface where untrusted implementation plan data is incorporated into subagent prompts. \n- Ingestion points: Implementation plan text and task descriptions are read from external files and interpolated into prompts in implementer-prompt.md and spec-reviewer-prompt.md. \n- Boundary markers: The prompts lack clear delimiters (such as XML tags or triple-backticks) or explicit instructions to the subagents to ignore potential instructions within the ingested task text. \n- Capability inventory: The subagents are granted significant capabilities, including the ability to modify the codebase, execute tests via shell commands, and commit changes. \n- Sanitization: There is no evidence of sanitization, filtering, or validation of the implementation plan content before it is processed by the subagents.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 12, 2026, 02:59 AM