feature-implement
Pass
Audited by Gen Agent Trust Hub on Nov 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied feature descriptions ($ARGUMENTS) and ingests codebase content via investigation tools to generate implementation plans and code changes.
- Ingestion points: User requests containing feature ideas or descriptions, and existing codebase files read during the research and planning phases.
- Boundary markers: The skill uses
<tasks>tags for user input but does not explicitly instruct the agent to ignore instructions embedded within the processed data. - Capability inventory: The workflow spawns multiple sub-agents (researcher, planner, code-reviewer, etc.), reads/writes files, and executes local Node.js and Python scripts.
- Sanitization: No evidence of sanitization or escaping of the user-provided $ARGUMENTS is mentioned before processing.
- [DYNAMIC_EXECUTION]: The skill invokes local Node.js and Python scripts (
.claude/scripts/ai-signal-scan.cjsand.claude/scripts/code_graph) as part of its automated research and verification phases.
Audit Metadata