Agent-facing skill for delegating workloads through the ray-hetzner Ray cluster on Hetzner Cloud.

Linux performance benchmarking skill for Claude Code — CPU, cache, branch, memory, ASM analysis with 7-dimension rubric

Leaf worker skill for ml-metaoptimization: analyzes experiment results against baselines and extracts learnings

Leaf worker skill for ml-metaoptimization: generates non-overlapping experiment proposals during the ideation phase

Leaf worker skill for ml-metaoptimization: implements experiment designs as code changes and patch artifacts

MCP server for Microsoft To Do via the Graph API. Multi-account, encrypted token storage, checklist support.

Audit test suites against TDD and behavior-first best practices. Produces severity-ordered findings and a prioritized improvement plan.

Leaf worker skill for ml-metaoptimization: ranks proposals and selects the winning experiment

Leaf worker skill for ml-metaoptimization: diagnoses failures during sanity checks and remote execution

Leaf worker skill for ml-metaoptimization: filters and curates the proposal pool during iteration rollover

Conservative, evidence-based test redundancy triage: classify tests as delete, merge, or keep using deselection, coverage, mutation, and branch-equivalence signals.

Leaf worker skill for ml-metaoptimization: designs concrete experiment batch specifications from winning proposals