Agent skills for Claude Desktop

Share a runnable prototype using the user's preferred deployment tool.

Local text-to-speech via sherpa-onnx (offline, no cloud)

Commit and push all local current-branch work, open a ready PR, babysit it, merge when clean, then create a fresh branch

Open Design's strategy memo: should it monetize the plugin registry now or hold — options, risks, and the recommendation.

Reference-free preference alignment, simpler than DPO.

Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that…

RL post-training for LLMs with Megatron and SGLang.

Create a Mastra project using create-mastra and smoke test the studio in Chrome using Chrome MCP server

Audio spectrograms/features (mel, chroma, MFCC) via CLI.

Creates specs before coding. Use when starting a new project, feature, or significant change and no specification exists yet. Use when…

Spring Boot architecture patterns, REST API design, layered services, data access, caching, async processing, and logging.

Spring Security best practices for authn/authz, validation, CSRF, secrets, headers, rate limiting, and dependency security in Java Spring…

Text-to-image generation, inpainting, and img2img.

Suggests manual context compaction at logical intervals to preserve context through task phases rather than arbitrary auto-compaction.

Stream-JSON chaining for multi-agent pipelines, data transformation, and sequential workflows

Provision SaaS services + sync creds via Stripe Projects.

Audit PostHog survey SDK features and version requirements

Fan out N parallel workers, drain them, and return one report.

Thread-safe data persistence in Swift using actors — in-memory cache with file-backed storage, eliminating data races by design.

Protocol-based dependency injection for testable Swift code — mock file system, network, and external APIs using focused protocols and…

SwiftUI Liquid Glass: implement, adopt, refactor, review correctness/perf/design.