43 tagged litellm, measured the same way as everything else here.
Browse within: RAG 13fastapi 13copilot 11deepseek 8openai 8self-hosted 8kimi 7model-router 7
Skill Claude CodeCodex
Maintain or assess a large, mature repository when a proposed or existing change may cross modules, platforms, clients, installers, configuration, security boundaries, tests, or documentation. Use for incoming-change adoption decisions, consequential bug fixes, refactors, dependency upgrades, compatibility work…
Skill Claude CodeCodex
Set up, verify, and troubleshoot spawning subagents on non-OpenAI models through codex-router. Use when the user wants Codex to delegate work to subagents, asks which models can act as subagents, says spawning a subagent fails or picks the wrong model, or wants to orchestrate parallel agents across routed providers.
Skill Claude CodeCodex
Create, list, read, message, wait on, fork, rename, archive, and pin Codex threads (sidebar tasks), plus automations and app navigation, using the app-native codexapp tools. Use when the session uses a custom (non-OpenAI) model, for example deepseek-v4-flash or mimo-v2.5, and the user asks to create a thread or a new…
Skill Claude CodeCodex
A guided design conversation for turning an idea for a feature, component, or system into an agreed implementation plan.
Skill Claude CodeCodex
Use when work should be delegated to Claude Code CLI, especially headless claude -p runs, automation scripts, CI jobs, resumable sessions, or requests to use Claude/Claude Code for a task.
Skill Claude CodeCodex
Review GitHub pull requests with evidence-backed, multi-perspective analysis and false-positive filtering. Use when the user asks to review, inspect, or check a GitHub pull request by number or URL. Default to reporting findings locally; publish comments, submit reviews, or approve only when the user explicitly…
Skill Claude CodeCodex
A set of rules and templates for packaging a Python MCP service in Docker. Docker packages an application and its dependencies into a repeatable container that can run on another machine.
Skill Claude CodeCodex
A set of rules and a Python structure for wrapping a service so AI clients can call its tools through MCP, an open protocol for connecting models to external tools. It uses server-sent events over HTTP for communication.
Skill Claude CodeCodex
Technical guidance for recognizing human actions in video without training a new machine-learning model or using an LLM. It describes a pipeline that samples video frames, estimates body poses, measures movement, and assigns labels with rules.
Skill Claude CodeCodex
Use when the user asks to build, create, or set up an app (a durable data app with a dashboard and background pipelines) — not a one-off widget or chart. Teaches how to delegate app-building to the app-builder sub-agent.
Skill Claude CodeCodex
Use when presenting information already in hand as a small, static structured panel in the conversation — status, results, comparisons, or a compact summary — rather than prose, a markdown table, a widget, or an app.
Skill Claude CodeCodex
Use when the user asks to render, build, visualise, mock up, or show a widget (chart, dashboard, card, table, graph) in the Mewbo Console. Teaches how to delegate the work to the st-widget-builder sub-agent instead of writing HTML or inline code yourself.
Skill Claude CodeCodex
Use when testing openclient-llm against a real LiteLLM server, OpenAI-compatible endpoint, virtual key, streaming SSE, tool calling, or model integration.
Skill Claude CodeCodex
Use when adding unit tests, improving test coverage, creating XCTest cases, or adding test mocks for existing openclient-llm Swift code.
Skill Claude CodeCodex
Use when building, running, testing, linting, or verifying the openclient-llm Xcode project, including fixing compiler, test, or SwiftLint failures.
Skill Claude CodeCodex
Use the delegate MCP tool to offload replaceable grunt work (boilerplate code, mechanical refactors, formatting, translation, summarizing long documents) to cheap models, so you keep your premium tokens for judgment work. Trigger whenever a subtask is mechanical, well-specified, and doesn't need your full reasoning.