AI & LLM Integration for Claude Code

Azure OpenAI Service — model deployments, fine-tuning, content filtering, prompt engineering, batch API, and quota management with az cognitiveservices and REST API

Azure Static Web Apps — JAMstack/SPA hosting with built-in auth, API backends, PR preview environments, and staticwebapp.config.json management

Copilot Studio — design bot topics, author trigger phrases, configure generative AI orchestration, and publish chatbots

Dynamics 365 Sales and Customer Service via Dataverse Web API — leads, opportunities, accounts, contacts, cases, SLAs, queues, pipeline reporting, and CRM workflow automation

Microsoft Fabric AI and operations agents - anomaly detector, data agent, operations agent, ontology, and digital twin builder workflows with preview guardrails.

Microsoft Fabric Data Science — ML experiments, model training, MLflow tracking, PREDICT function, and semantic link integration

Microsoft Fabric graph and geospatial analytics - graph model, graph queryset, map, and exploration workflows with preview guardrails.

KDB.AI vector database — schema design, hybrid search, and integration patterns for AI applications

Research Microsoft APIs, scaffold new plugins, extend existing ones, and audit marketplace coverage

Convert any API, GitHub repo, or OpenAPI spec into a production-ready MCP server using AI — analyze, generate, test, and rollback from Claude Code

Microsoft Graph + Entra + Office file ops MCP for Claude. Certificate-based app-only auth via Azure Key Vault + delegated auth with persistent token cache. Covers Outlook, Teams, SharePoint, OneDrive, Planner, ToDo, OneNote, Users, Groups, Entra admin, Excel/Word/PowerPoint/Loop file ops, Search, Sites.

Consult GPT-5 Pro for alternative perspectives; supports context injection, web search, and configurable models

Give Claude Code agents the ability to send themselves continuation messages mid-session — metacognition, not repetition

11 reusable skills for AI agents (planning, code-writing, testing, etc.) plus a /skillfold compiler command

Custom Teams app development — manifest v1.25, M365 Agents Toolkit, Adaptive Cards, message extensions, meeting apps, Custom Engine Agents, Agent 365 blueprints, workflow bots, notification hubs, Copilot plugins, Teams SDK migration, and advanced meeting experiences with Live Share

36-command Claude Code skill for X/Twitter — post, search, engage, moderate, discover communities, all from your terminal

AI personality test — get roasted based on how you talk to AI

Skill for orchestrating independent Claude Code sessions in cmux terminal — split panes, agent monitoring, browser automation, sidebar metadata, notifications, and full CLI coverage.

Codex CLI reflector — independent critic, oracle, and metacognition layer using OpenAI Codex as second-model reviewer

Neural network interpretability with nnsight - includes skills for tracing, logit lens, activation patching, attribution patching, causal tracing, model steering, and remote execution

DeFi toolkit for AI agents — swap tokens, check balances, manage wallets, and configure safety guardrails on the Sui blockchain via the fence CLI.

Re-prompts Claude Code when it stops prematurely — catches PERMISSION-SEEKING, STOPPED-WITH-TODOS, and FALSE-COMPLETE failure modes (78% of real agent stops are premature), and injects targeted recovery instructions via the Stop hook.

Evaluate AI skills across model tiers with blind testing and refinement recommendations

Bidirectional bridge between Claude Code and external AI coding agents. Exposes Claude Code as an A2A server (Gemini CLI, any A2A client) and as an ACP agent (OpenClaw, Zed, VS Code); routes outbound tool calls to Codex, OpenClaw, and Hermes adapters through a shared daemon with per-room task logs and verification-artifact support.

ElevenLabs Claude Code plugins enabling voice-based interactions and audio processing for AI workflows.

A portable GTD skill for Claude Code: capture / clarify / organize / engage / review over plain-Markdown lists in memory/gtd/. The bundled skill auto-activates; /gtd and /gtd-* commands are included. Calendar writes are confirmation-gated and fail-closed.

Teach your AI coding agent the MASA (Modular Agentic Semantic Architecture) framework

Automatically generate Claude plugins from user prompts. Supports plugin creation, documentation, and testing.

Create formal, verifiable proofs of claims with machine-checkable reasoning

Accelerated speech-to-text transcription using optimized Whisper model for faster audio processing and transcription.

TRTC SDK integration assistant with knowledge base

> Edit existing video on RunComfy — this skill is a smart router that matches the user's intent to the right edit model in the RunComfy catalog. Picks Wan 2.7 Edit-Video (general restyle / background swap / packaging swap, identity + motion preservation), Kling 2.6 Pro Motion Control (transfer precise motion from a reference video to a target character), or Lucy Edit Restyle (lightweight identity-stable restyle / outfit swap). Bundles each model's documented prompting patterns so the skill gets

Every Claude-compatible skill — all 47 skills plus the internet-researcher agents. Heaviest context cost; prefer a themed bundle or single skill.

Submit slide classification training jobs on Azure GPU compute with full pipeline or quick mode.

Auditing AI/ML personnel assessments for fairness, bias, validity, and utility, using the 12-component psychological-audit framework of Landers & Behrend (2023), 'Auditing the AI Auditors.'

Evaluating AI-based selection tools against scientific, legal, and ethical standards (Uniform Guidelines, Title VII, APA Ethics Code), per Tippins, Oswald & McPhail (2021).

Dream-briefing for AI coding agents: a cognitive checkpoint at session end, a 'while you were away' briefing at session start.

Persistent AI project memory — restore session context, recall & save knowledge, setup, and an MCP smoke test