feature-implement

Pass

Audited by Gen Agent Trust Hub on Nov 4, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from project documents and user feature descriptions which can be used to influence the behavior of the subagents employed in the workflow. Ingestion points include docs/project-config.json, docs/specs/, plans/, and user-provided task inputs. Capability inventory includes spawning multiple subagents (researcher, planner, tester) and executing project-local scripts via Node.js and Python for codebase analysis. Boundary markers such as the 'Goal Contract' and success criteria are used to focus implementation, but the skill lacks explicit delimiters or instructions to ignore embedded commands within the files it processes. Sanitization of the ingested file content is not explicitly defined in the skill's protocols.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 4, 2026, 10:12 AM