autoplan
Pass
Audited by Gen Agent Trust Hub on Nov 16, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the Bash tool to run various vendor-provided utility scripts and binary tools for session management and review automation. These operations are consistent with the skill's core functionality.\n- [DYNAMIC_EXECUTION]: The skill dispatches subagents with context-specific prompts generated at runtime based on development plan snapshots, enabling independent multi-phase reviews.\n- [PROMPT_INJECTION]: Instructions include explicit mitigations against indirect prompt injection by instructing subagents to ignore foreign instructions within repository code during analysis.\n- [EXTERNAL_DOWNLOADS]: The skill records telemetry using vendor-provided tools and suggests installation of official dependencies, all targeting the author's official infrastructure.\n- [SAFE]: The skill implements a publication guard that monitors the session transcript locally to ensure process integrity. This ensures that the agent follows the mandated review methodology without exposing data to unauthorized parties.
Audit Metadata