"JSON Web Tokens" matching MCP servers:

Enables Claude Code to run OpenAI Codex (GPT-5) tasks asynchronously in the background, filtering out thinking logs to save 95% context tokens and allowing parallel execution.

Wraps MiniMax AI as an autonomous code executor for Claude Code, offloading coding tasks to save Claude subscription tokens.

A lightweight MCP bridge that lets Claude Code delegate heavy tasks to the Antigravity CLI - saving context window and tokens for what matters.

Provides a durable memory layer for coding agents like Claude Code and Codex by indexing codebases and enabling RAG queries, reducing rediscovery tokens and providing senior-engineer orientation.

MCP server for the AIMFP paradigm, enabling AI assistants to manage functional programming projects with persistent database state, directive compliance, and task tracking via JSON-RPC tools.

Specialized MCP server for coding workflows with web search and image analysis tools.

A domain-agnostic MCP server for autonomous experimentation, generalizing Karpathy's autoresearch pattern into a reusable server that any AI agent can drive, pointed at any domain defined by a JSON configuration.

STOP UI SLOP. Gives coding agents searchable evidence from 800,000+ real web and iOS screens, design contracts, and a hard UI finish gate.

Generates compiled Flutter apps from JSON Product Requirements Documents. Exposes endpoints for PRD validation, schema, payment terms, and build status.

Enables AI coding agents to save task context as a validated JSON document and later restore it as a concise briefing with the objective, status, relevant files and focus lines, decisions, and prioritized next steps.

MCP server that exposes independent OpenCode AI coding agents to upstream agents with multi-workspace concurrent routing, enabling session management and message exchange across multiple repositories via stdio JSON-RPC.

Enables ChatGPT Web Developer Mode to interact with local repositories through MCP, providing tools for file editing, shell execution, Git worktrees, semantic navigation, and checkpoints.

Enables AI coding agents to conserve tokens by enforcing rules like delegating heavy work to subagents, checkpointing before token limits, and grepping before reading files.

An MCP server that wraps the Claude Agent SDK, enabling Claude-powered queries, coding tasks, web search, and customizable agent execution using OAuth without an API key.

Enables MCP clients to read a project's component tree and design tokens, edit React/TSX canvas frames, set props, apply auto layout and variants, add shader backgrounds, and capture screenshots, so agents can make design changes that are written back as reviewable code diffs and pull requests.

Provides a file-based, git-friendly memory store for coding agents to persist and recall project-specific lessons, bugs, and conventions via JSON files in the repository.

Exposes a codebase's symbol graph and symbol-aware editing tools to AI agents, enabling targeted symbol lookup, impact analysis, and atomic multi-file edits with reduced context tokens.

Enables Codex to start the local DeepSeek Harness Web UI, delegate coding tasks into visible sessions, and review workspace changes.

An MCP server implementing the TRUE Recursive Language Model technique for managing large context windows in Claude Code, enabling analysis of codebases beyond 200k tokens by storing content as variables and using LLM-generated code for search and analysis.

Enables AI coding agents to read and write a user's Figma file through the Figma Plugin API, offline and privately, without API tokens or rate limits.