aidlc

Pass

Audited by Gen Agent Trust Hub on Jul 20, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill performs legitimate filesystem operations to manage project state, including creating manifests, audit logs, and platform-specific entry points (shims like .claude/CLAUDE.md). These actions are performed using built-in platform tools such as fsWrite or Edit.
  • [SAFE]: The orchestrator uses a modular 'Skill Dispatch' pattern to load phase-specific instructions from local disk. While this involves a context override pattern ('disregard any prior... instructions'), it is a core architectural feature used to transition between defined workflow steps (e.g., from 'requirements' to 'design').
  • [SAFE]: The skill includes maintenance and recovery tools (doctor, repair, upgrade) designed to verify installation health, rebuild state from disk artifacts, and migrate legacy project structures. These utilities facilitate environment consistency and reliability without external dependencies.
  • [SAFE]: The tool adheres to a local-first policy for data processing, reading from and writing to specific project directories (.aidlc/specs, .aidlc/workflow, etc.) without any observed network activity or external data exfiltration patterns.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 20, 2026, 08:00 AM
Security Audit — agent-trust-hub — aidlc