Quality Gates
Validate search, target selection, plan, and execute changes with repeatable quality gates.
Benchmarks
Measure retrieval, selection, planning, and context reduction quality.
BFCL-Style Evaluation
Document how graph-tool-call can be evaluated with tool-call benchmark methodology.
XGEN Scale Gates
Replay large API collection snapshots to protect enterprise catalog quality.
Quality Lab
Run search, plan, and execute cases against a collection before enabling production usage.
Release Gates
Choose validation depth for local development, release candidates, and public claims.