ln-26-architecture-migration-planner

Pass

Audited by Gen Agent Trust Hub on Oct 26, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill is designed for planning and documentation. Its instructions explicitly restrict the agent from executing migrations, editing product code, or authorizing deliveries, maintaining a read-only posture relative to the core system.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it ingests untrusted external data during its planning phases.
  • Ingestion points: The skill reads repository instructions, deployment configurations, and external vendor documentation to verify version-dependent assumptions (Checklist 1 and Tool Routing).
  • Boundary markers: No explicit delimiters or instructions to ignore embedded prompts in external documentation are present.
  • Capability inventory: The skill is authorized to read repository files and use discovery tools (schema tools, telemetry). Write access is strictly limited to the migration plan artifact (e.g., docs/architecture/migration-plan.md).
  • Sanitization: The skill lacks explicit sanitization for external content, though it includes a security-positive pattern of summarizing bounded results while keeping full tool output outside the context window.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 26, 2026, 10:49 AM