AI Engineering

MCP Apps Builder scaffolds, modifies, migrates, and reviews TypeScript MCP servers and interactive MCP Apps using mcp-use v2. Use it when building tools, resources, prompts, Views, React host interactions, OAuth, middleware, Inspector workflows, and package-boundary migrations.

MCP Builder builds, modifies, and debugs fully typed TypeScript MCP servers with mcp-use, covering tools, resources, prompts, agent skills, OAuth adapters, middleware, and production transports.

OpenAPI to MCP builds and deploys an MCP server from an OpenAPI or Swagger spec so an LLM can call a REST API. Use it to expose an existing HTTP API as LLM-callable tools with proper schemas and auth.

Budget Negotiator enables an L3 autonomous loop to request a token budget increase when near its cap and high-priority work remains. It calculates ROI from run logs and escalates a tagged request for human approval.

Loop Constraints reads loop-constraints.md at the start of every run and enforces all rules before triage or actions. Use it to apply binding guardrails such as denylist paths, push/merge checks, and test requirements.

Loop Triage reviews recent CI failures, commits, issues, and conversations, then produces a concise, prioritized findings report for an AI loop to act on. Use it to separate high-priority work from noise and persist state between runs.

Handoff writes or updates a HANDOFF.md document so the next AI agent with fresh context can continue the work. Use it to record goals, progress, what worked, what didn't, and next steps before handing off a coding session.

Quarter Clone clones the last quarter of the current conversation to reduce token usage while preserving recent work. Use it when you want to discard old context and continue from the most recent portion.

Review CLAUDE.md analyzes recent Claude Code conversations to find improvements for global and project-level CLAUDE.md files. Use it to reinforce violated instructions, add project-specific patterns, and identify outdated guidance.

Using AO catalogs the AO (Agent Orchestrator) CLI for spawning workers, managing sessions and projects, sending messages, controlling the shared browser, previewing pages, and daemon control.

Antigravity SDK E2E Dev spins up a live local Omnigent server and exercises the Antigravity (Gemini) SDK harness end-to-end, including building agents, running turns, and smoke-testing. Use it when developing, testing, or debugging the Antigravity harness.

Build Omnigent Agents offers patterns and templates for creating valid Omnigent agent directories, covering minimal, web-search, harness, MCP, and multi-agent configurations. Use it when generating files for an Omnigent agent.