agent-code-goal-planner
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill documentation and workflow examples utilize
npx claude-flow, which fetches and executes theclaude-flowpackage from the standard npm registry as part of its intended development workflow. - [COMMAND_EXECUTION]: The skill uses the
npxrunner to execute various shell-based sub-commands for TDD, architecture planning, and feature implementation tasks. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest external user requirements and process them through automated planning and execution cycles. 1. Ingestion points: User-provided software requirements and goals defined in SKILL.md workflow. 2. Boundary markers: Absent. 3. Capability inventory: Shell command execution via
npxdescribed in SKILL.md. 4. Sanitization: Absent. This surface is evaluated as a low-risk architectural characteristic common to development assistants.
Audit Metadata