AgentX benchmarks, recipes, and infrastructure

Run AgentX throughput with 1 warmup request per lane and a 20-minute profile; not reusable

Expand eval selection to every fixed-sequence config

Something isn't working

Task for Claude to handle

TEST: mock claude-task for local script verification — do not action

CODEOWNER checklist passed once; retained across PR commits

Pull requests that update a dependency file

Automated Docker image update notifications

TEST: mock data from local docker-tag-monitor.sh runs — do not action

Improvements or additions to documentation

This issue or pull request already exists

New feature or request

Suppress throughput and run only eval jobs; combine with all-evals to expand selection

Full sweep with canary gate; matrix jobs run to completion despite failures

Full sweep with canary gate; first failure cancels the rest of that matrix (recommended)

Pull requests that update GitHub Actions code

Good for newcomers

Extra attention is needed