What problem does it solve?
It prevents an AI agent from running full plans on high-risk or system-impacting work that needs extra architectural scrutiny, reducing rework and missed constraints.
Core Features & Use Cases
- Quantifies task complexity using a weighted scoring model across scope, architectural impact, technical risk, dependencies, and uncertainty.
- Auto-detects critical scenarios and forces complexity to at least 80% when high-leverage triggers are present (e.g., auth changes, schema migrations, breaking API contracts, security-critical features).
- Routes workflow rigorously by selecting Standard, Enhanced, or Critical (deep architectural review) thresholds, and generating a decision report for provenance logging.
Use cases: before plan generation in a multi-phase workflow, when deciding the appropriate validation depth, or when determining whether a task requires a human/dedicated architecture review.
Quick Start
Use the complexity-router skill to score and route the next task before plan generation, ensuring anything at or above 80% triggers deep architectural review.
