AI & Agents Agent Skills for OpenAI Agents SDK

Felo SuperAgent API: AI conversation with real-time SSE streaming on a persistent LiveDoc canvas.

Few-shot example generation and optimization for improved LLM performance

Use this skill when building real-time, bidirectional streaming applications with the Gemini Live API.

Transforms vague UI ideas into polished, Stitch-optimized prompts.

Multi-perspective deliberation on a question.

Create new skills, modify and improve existing skills, and measure skill performance.

Route tasks via the 3-tier model selector and learned patterns; emits a routing rationale via hooks_explain

Evaluates LLMs across 60+ academic benchmarks (MMLU, HumanEval, GSM8K, TruthfulQA, HellaSwag).

Evaluate-Loop Step 4: EVALUATE EXECUTION.

Configure MCP servers for n8n development.

Integrate Mem0 into an existing repository using a goal-driven, TDD pipeline that writes failing tests before any implementation.

Use when opening a RoundTable proposal in the active Markdown-backed Agent Team Mission.

Use when starting or continuing a Markdown-only Agent Team Mission with Claude Code or Codex background Agents and no Mexus server.

This skill should be used when the user asks to "design multi-agent system", "implement supervisor pattern", "create swarm architecture"…

Evaluates LLMs across 100+ benchmarks from 18+ harnesses (MMLU, HumanEval, GSM8K, safety, VLM) with multi-backend execution.

Automate OpenAI API operations -- generate responses with multimodal and structured output support, create embeddings, generate images…

Scan a repository for Opus-4.6-era patterns that break or degrade on Opus 4.7 — fixed-budget Extended Thinking parameters, retired model…

Automatically converge from goal to A-grade Seed and execute it

Use this skill for \"initialize project\", \"set up project\", \"scaffold project\", \"create AGENTS.md\", \"create CLAUDE.md\", \"set up…

Multi-mode code and skill review for controlling drift as capability surface grows.

Build or audit an evidence-driven SGLang Day-0 support program for a new LLM, VLM, MoE, hybrid-attention, or speculative-decoding model.