MCP Servers tools for opencode

Semi-automated research assistant for academic research and software development. Supports Claude Code, Codex CLI, Kimi Code CLI, and OpenCode across ideation, coding, experiments, writing, and publication.

Reverse engineer any website into a typed API client. Works with Claude Code, Codex, Cursor and OpenCode.

A persistent, Git-versioned map of your whole codebase and database schema that coding agents read before they touch anything. Local-first MCP server + CLI in Go: a governed repository index of code knowledge that gives Claude Code, Codex, Cursor, and opencode long-term context, memory, and code intelligence.

Indexes coding-agent sessions already written to disk and serves them back over MCP with no LLM calls, embeddings, or API keys. The zero-dependency binary solves the memory cold-start problem for cross-session persistence and demonstrates that agent memory can be built on existing filesystem traces rather than a separate learning pipeline

Bash for AI agents on Windows via deterministic bash-to-PowerShell translation, without WSL or a VM, with GNU-style output and persistent cwd and env

Open-source, multi-platform harness for AI agents - a native, multi-threaded operator's console (JVM, not Electron) to run Claude Code, Codex, Gemini or OpenCode with a real browser, terminal, editor, secrets & 200+ MCP tools. Built for enterprises, science & research.

Accessibility-first browser & computer use for AI agents — one MCP server for every harness (Codex, Claude Code, OpenCode, Antigravity). Instruments the DOM/accessibility tree via CDP instead of screenshot loops: diffed snapshots, stable element refs, trusted input, typed errors. Zero dependencies

Memory system for coding agents: backfill ranks earlier records as candidate causes of a fresh failure even when they share no vocabulary with it, the composed graph records which memories were used together and surfaces never-tried combinations, retrieval decays on an FSRS-6 schedule, and receipts fail closed after compaction. Single Rust binary, local only. Install: gemini extensions install htt

Open-source, channel-native agent gateway for Slack. Route team threads to Claude Code, Codex, OpenCode, Docker, HTTP agents, and custom CLIs with policy, approvals, memory, audit logs, and artifacts.

Free, open-source MCP server that gives AI coding agents (Claude Code, Codex, Cursor, OpenCode, ZCode…) design inspiration from award-winning websites — search with real screenshots, pull design DNA, capture and record live sites. A Mobbin alternative. No API key.

An effective and efficient coding-agent setup in under 30 seconds: best tools, unified instructions, any OS.

Cross-model, local-first episodic memory engine for Claude Code, Cursor, Codex, and Antigravity. SQLite FTS5 (BM25) + 8MB static model2vec CPU embeddings via Reciprocal Rank Fusion; stdio MCP server for durable decisions, budget-capped boot context, and hybrid recall. Zero cloud, zero telemetry

Website audit MCP server for coding agents: run audits, read reports, and fix findings from Gemini CLI or any MCP client. 260+ rules across SEO, performance, security, accessibility, and agent experience, with fixes mapped to source. MIT CLI with a local stdio server (squirrel mcp) or hosted Remote MCP with OAuth: gemini mcp add --transport http squirrelscan https://mcp.squirrelscan.com/mcp

ChatGPT的MCP 桥接工具，赋予网页版ChatGPT本地读写能力，并将ChatGPT连接至本机codex、agy、dsh、opencode，作为大脑调用本机agent。Connect ChatGPT to local projects and coding agents via MCP. Run Codex, OpenCode, DeepSeek Harness & Antigravity with approvals, live output and session control.

Persistent, provider-neutral memory for Codex, Claude Code, OpenCode, Pi, and MCP coding agents.

Reusable prompt templates with multi-step chains, quality gates and reasoning frameworks, exportable as native skills for coding agents

Turn raw clips into a finished cut through chat

MCP server that gives AI assistants a shared view of your terminal session. Debug CLIs and TUI apps in real-time or let agents drive terminal-based tools autonomously. Works with Claude Code, Codex, Cursor, opencode, and Copilot

Indian stock market data for Gemini CLI: quotes, quarterly financials, shareholding, a 326-ratio screener, FII/DII flows, option chains and company filings for about 8,200 NSE and BSE stocks. Remote Streamable HTTP with OAuth or a free personal token, or local stdio via npx -y tapetide-mcp; free plan available. MIT. Install: gemini extensions install https://github.com/Tapetide-hq/nse-bse-indian-s

AI council: query CLI agents (Claude Code, Codex, Gemini and OpenCode) in parallel with deliberation rounds

Install a large security toolset and discover and run it through an authorization-gated server for CTF, pentesting, bug bounty and DFIR

Persistent coding memory across sessions, machines and teammates, with memory search, optional sync, project sharing and automatic capture

Local-first, source-grounded project memory for Gemini CLI and other coding agents. Provides bounded recall, citations, atomic changesets, an installable Agent Skill, and a read-only stdio MCP server (lwc serve --mcp). Apache-2.0

Checks what local coding agents such as Claude Code, Codex and Cursor really did: files changed, commands run and test results read from test output, not the agent's word. Read-only

Remote MCP server for browsing model catalogs, checking pricing, and creating image, video, music, audio, and other model API tasks through RunAPI. Works with Gemini CLI: gemini mcp add --transport http runapi https://mcp.runapi.ai/mcp

by roampal-ai - Outcome-based persistent memory MCP server for Claude Code and OpenCode. Good advice promoted, bad advice demoted. pip install roampal

Persistent memory for AI coding tools with semantic code search, knowledge graphs, impact analysis and decision tracking

Long-term memory for coding agents as a tree of plain local files: searches before answering, keeps changed facts as history and flags them OUTDATED, and returns only the 200–900 tokens a question needs

Offline RPG Maker MV assistant: generate maps, edit the database and events, validate broken references, debug events and search a project semantically

Search developer infrastructure deals, free tiers and startup programs, compare vendors, plan stacks and track pricing changes

Reasoning Harness for agentic AI: 4 MCP tools (reasoning, code, anti-deception, memory) over 679 engineered cognitive operations. Each call returns a structured scaffold the calling LLM ingests before its first token. Catches sycophancy, hallucination, and reasoning decay before they emerge. Companion skills/ directory ships SKILL.md files for autonomous routing in Claude Code. MIT, free tier 100

Reads local Claude Code, Codex and OpenCode usage logs: token totals, daily and hourly activity, per-agent and per-model breakdowns and a year-in-review recap

Provider-neutral web search aggregating model grounding (Gemini, Grok, DeepSeek and others) and search APIs such as Exa, Brave, Perplexity and Tavily

Local-first learning substrate + MCP server for Claude Code, Codex, OpenCode, and Cursor. Builds a Work Model from what actually worked (graded by real outcomes) and predicts the better path, so the same job gets faster and lands better every run

Delegates tasks between local coding-agent CLIs, including Claude Code, Codex, OpenCode and Antigravity, and returns results to the calling agent

Local-first safety and verification layer for coding agents: workspace confinement, command allowlists, scope-violation detection and auditable task evidence

Local-first persistent memory for AI coding agents: plain Markdown as the source of truth with SQLite FTS5 search, one memory shared across VS Code, Visual Studio, Cursor, opencode and Claude Code, and team sharing via git PRs

Bilingual (English/Spanish) dual-memory system on PostgreSQL, pgvector and NVIDIA NIM with progressive retrieval, deduplication, checkpoints and auto-repair

A Linux system monitoring MCP server providing real-time CPU, memory, disk, network, processes, Docker, systemd, security audit, and hardware info via STDIO transport. Install: download from GitHub Releases or go install github.com/Mohabdo21/linux-mcp@latest

Local execution memory for coding agents: injects what worked and known fixes before similar tasks and advises whether to continue, pivot or stop

Local browser MCP for coding-agent QA with verified actions, screenshots, read-only network and console inspection, and interruption recovery warnings

Cross-agent failure intelligence: before retrying a failed tool call, check whether other agents hit the same failure and which recovery worked

Shared, serverless, conflict-free memory for AI agents: a multi-machine drop-in for the MCP memory knowledge graph, converging per-node shards with no server

Manage Substack posts, notes, comments, replies, reactions and restacks, with a propose-then-confirm reply flow, audit log and dry-run by default

ADB-based Android access: read-only logcat, UI hierarchy, package listing and allowlisted shell commands

Feature-complete Discord MCP server with 117 native tools, dual user/bot account support, anti-abuse human pacing, and dynamic aesthetic symbol discovery

Localhost review board where different coding agents (Claude Code, Codex, OpenCode, Trae, ZCode) review each other's work: objections must state flip conditions, revisions clear all verdicts, resolution requires the whole active quorum, and every vote is append-only audited. Ships its own audited self-review history as evidence. Install: pipx install mcp-review-board