MCP servers that ship as code — a repository or a package you install and run locally (usually over stdio). They have no hosted endpoint, so we cannot probe or grade them: instead each one shows its repository, install command, license, stars and when it was last updated.
Collected from GitHub repositories and npm, PyPI, Cargo, Go, Maven, NuGet and RubyGems packages tagged as MCP servers, the official MCP Registry, Docker's MCP catalog, ModelScope, Baidu MCP World, Cline and other public directories; one entry per repository. A server that also offers a hosted endpoint is listed under Remote instead. These entries are not counted in our reachability, verification or readiness statistics.
-
agent-replay-mcpgithub.com/mdfifty50-boop/agent-replay-mcpupdated this year
MCP server for agent session recording and replay — debug non-deterministic agent behavior with session comparison and divergence detection
npm -
Python FastAPI recorder for MCP session replay, approval history, evidence chains, and operator-facing audit posture.
MIT -
A concurrent browser MCP server with session recording & replay - supports multiple parallel browser instances and captures all actions as JSON
npm Apache-2.0 -
claude-code-vcrgithub.com/PaoloJN/claude-code-vcrupdated this year
Local-only MCP server for replaying and searching Claude Code session logs
MIT ★ 1 -
@supastory/mcp-servergithub.com/supastory/supastoryupdated this year
SupaStory MCP server — bring AI-powered session replay insights directly into your editor
npm SEE LICENSE IN LICENSE -
MCP server for frontend UI debugging — component inspection, visual regression, a11y auditing, and session replay via CDP
npm MIT -
Open-source, privacy-first analytics for 3D scenes — view-direction & click heatmaps, mesh interactions, performance, and session replay. Self-hosted, multi-engine.
npm Apache-2.0 ★ 3 -
Hosted MCP server for PulsePanda: analytics, session replay, funnels, Google Search Console + Bing data, surveys, feedback and help desk for Claude, ChatGPT, Cursor and any MCP client
-
A session replay engine an AI agent can interrogate. 17 WebMCP tools: it binary-searches the timeline for the millisecond a bug began, then tells you what it cannot see and asks you to look.
MIT ★ 1 -
rrweb-stackgithub.com/Cubenest/rrweb-stackupdated this year
Two OSS products on one rrweb substrate: tracelane (self-contained HTML test-failure replays for WDIO/Playwright/Cypress) + peek (local-first browser-session forensics + repro for AI coding agents, over MCP). No SaaS, no telemetry, local-first.
npm Apache-2.0 ★ 9 -
mcp-session-memorygithub.com/zavora-ai/mcp-session-memoryupdated this year
Session Memory MCP server for ADK-Rust Enterprise — typed session state, scoped memory recall, replay snapshots, and resumable workflows
cargo Apache-2.0 -
Open-source, self-hosted observability for AI agents: OTLP ingestion, tracing SDK, sessions, cost analytics, datasets and LLM-as-judge experiments, alerts, replay, and an MCP server. SQLite or PostgreSQL.
pypi MIT -
waspholegithub.com/riza/waspholeupdated this year
Open-source MCP honeypot with fake tools, canary credentials, session replay, and alerting.
Apache-2.0 ★ 5 -
Terminal automation for AI agents: PTY sessions, image-aware screenshots, visual regression, recording/replay, and MCP.
MIT ★ 1 -
MCP server for PostHog, cloud or self-hosted: HogQL, event and property definitions, session replays and the REST API.
npm MIT -
winshotsgithub.com/SofianeBel/winshotsupdated this year
Open-source, local-first Windows capture and context bridge for Codex with MCP, OCR, visual sessions, Instant Replay, and Agent Watch.
MIT -
adtivity-mcpupdated this year
Adtivity MCP server — install Adtivity analytics, session replay, error tracking and churn prediction into any codebase via Claude Code or Cursor in under 10 minutes
npm MIT -
Model Context Protocol (MCP) server for Webeyez. Access Webeyez session replays, JavaScript errors, conversion funnels, and performance diagnostics directly from any MCP-compatible AI client.
npm ISC -
Agent-first browser automation runtime — deterministic Chromium sessions with replay-equal hash chains, MCP-native tools, and a content-addressed action store.
Apache-2.0 ★ 1 -
qorrelate-mcp-servergithub.com/sethmichaelking/arqiveupdated this year
MCP server for Qorrelate observability platform - 123 tools for AI agents to manage logs, traces, metrics, session replays, and alerts
npm MIT -
mcp-engramgithub.com/hugfeature/engramupdated this year
Durable Agent Runtime — append-only event log, replay-based recovery, cross-session task continuity, agent handoff. Local-first, zero cloud dependency.
pypi MIT -
Free, open, account-free communication utility for agents — signed sessions, AEAD, replay protection, and MCP
pypi MIT -
MCP bridge for the running DeepSeek Harness desktop: 26 tools over a loopback MCP endpoint so Antigravity, Claude Code, Codex or Cursor drive the same live sessions, answer approvals and questions from the MCP side, keep per-session interaction modes, and use a reconnecting stdio proxy with a no-duplicate replay policy.
MIT -
Control plane for browser fleets. One endpoint, any provider: failover, persistent profiles, session replay, REST API, MCP server, dashboard. OpenRouter for cloud browsers.
go MIT ★ 19 -
See the bugs your users hit. Let your coding agent fix them. Self-hosted session replay with a CLI and MCP server for Claude Code and Codex.
MIT -
Local-first, self-hosted AI agent runtime and MCP bridge with sandboxed sessions, memory, credentials, audit/replay, and a local Console.
oci Apache-2.0 ★ 643 -
A live, local-first dashboard for Claude Code sessions: project graph of agents, tokens, tool feed, replay, alerts. Plugin + terminal UI + MCP server. Includes the AOS-1 agent orchestration spec.
MIT ★ 104 -
@neurcode-ai/mcp-servergithub.com/sujit-jaunjal/neurcodeupdated this year
Model Context Protocol server for the Neurcode deterministic governance runtime. Local-first. Zero-install via npx. Replay-stable. Source-free lifecycle ingress for governed AI coding sessions.
npm MIT -
vibepilot-mcpgithub.com/ravindra240385/vibepilotupdated this year
AI-native project management MCP server — 17 consolidated tools, 17 resources, 8 persona prompts for vibe coders. Features, sprints, decisions, knowledge base, session replay, conflict detection, and more.
npm MIT -
lucarne-corpus-mcpgithub.com/volter-ai/lucarneupdated this year
Optional thin stdio MCP bin: read-only queries over a lucarne-records store. Never fetches, replays, or opens a browser — a miss returns a structured not_captured result telling the agent to browse to it in-session.
npm MIT -
probagithub.com/evotech-bg/probaupdated this year
The QA workbench for the agent era: drive web/API/DB through MCP and own the tests you get back (Gherkin <-> JSON <-> Playwright). Replay evidence, failure clips, self-healing, cross-session memory. Local-first, MIT.
MIT -
Turn a user-reported bug into a runnable Playwright regression test — privacy-safe, no session replay, no PII captured. An open-source, self-hostable evidence layer for regulated and quality-focused teams.
npm Apache-2.0 -
prime-agent-posthog-mcpgithub.com/gbrlpzz/prime-agent-posthog-mcpupdated this year
Official PostHog MCP integration for Prime Agent: talks to PostHog's own MCP server (mcp.posthog.com) with a project-scoped personal API key, giving the agent analytics, funnels, path/user-flow analysis, trends, dashboards, and session replay out of the box (MIT).
MIT ★ 2 -
StrawManus - a browser-control agent that fuses Manus's flat 12-tool action layer with Strawberry's session, natural-language selector resolution and skill-reuse layer. Runs in your real logged-in browser, self-heals when sites change, and replays successful tasks as skills. Includes an MCP server.
-
Cross-platform Chrome MCP that runs on a persistent clone of your real Chrome profile (session/logins carried over) with real-time request interception, response modification, network capture, HAR recording/replay, accessibility testing, stealth mode, con
npm MIT -
Audit trail database for AI agents. Tamper-evident, checksum-backed decision logs with session replay and time-travel debugging to support EU AI Act Article 12 record-keeping. Drift detection, MCP, Python & TypeScript SDKs. Self-host or cloud.
npm AGPL-3.0 ★ 162 -
Real-time Linux perf profiling in your browser. A tiny static C agent streams perf data from any ARM or x86 Linux device, big-endian included, and a Python server renders live flame graphs, function tables and line-level annotated source. Session replay, perf.data import, and an MCP server for AI agents.
MIT ★ 1 -
Automate Blender 3D modeling with AI via MCP — 93 tools for modeling, sculpting, architecture/MEP, materials, and 3D-print prep. Includes Blender Studio: a visual record/replay session editor with a multi-provider AI assistant (Claude, Gemini, OpenRouter) — drive it from n8n, Claude Code, or standalone.
pypi MIT ★ 86 -
Local cost analyzer for AI coding agents: see where tokens and AI Credits go in Claude Code, GitHub Copilot Chat, pi and mitmproxy sessions. Turn-by-turn cost, prompt caching explained, cost by repository and project, replay on another model, tool-call audit, MCP server with interactive views. Also a VS Code extension: AI Credits Analyzer.
MIT ★ 1 -
@nomadop/session-watchergithub.com/nomadop/session-watcherupdated this year
Local Claude Code context-cost monitor, transcript replay, buckets, and handoff
npm MIT -
Session checkpoint and recovery for Claude Code, resume interrupted sessions, browse them in a dashboard/TUI, and export shareable HTML traces. Passive hooks, local SQLite, 5 MCP tools. Offline-first.
pypi MIT ★ 1 -
@malleon/mcpgithub.com/malleon-io/malleon-mcpupdated this year
MCP server for Malleon Replay — diagnose bugs, create reference replays, and verify fixes via AI agents
npm SEE LICENSE IN LICENSE -
@mouseflow-utils/mouseflow-mcpupdated this year
MCP server for the Mouseflow public API
npm ISC -
MCP stdio server for inspecting and analyzing WebBlackbox archives.
npm MIT -
ghostplay-mcpgithub.com/gufao/ghostplay-mcpupdated this year
Ghostplay MCP server for Claude Code and other MCP clients
npm MIT -
Clintplay MCP server for Claude Code and other MCP clients
npm MIT -
Agent skills and the Convertlyft MCP server for Claude Code, Codex and Cursor: funnels, replays, errors, speed and SEO.
MIT ★ 1 -
Self-hosted observability for agentic debugging: traces, logs, replay, AI cost, Agent Action Graphs, MCP, profiles, and usage analytics in one evidence graph.
MIT ★ 2 -
bugtapeactive
BugTape CLI: connect Claude Code, Cursor, Codex or Grok to your production bugs over MCP. Replay, request, console and stack for every user-reported or silent failure.
npm Apache-2.0 -
UserProof for Claude Code: real proof of why users leave your site, and whether your fix worked.
npm MIT -
Voice-anchored causal evidence recorder for AI coding agents — talk, point, and get an evidence package your agent can act on. 语音锚定的因果证据记录仪。
npm MIT ★ 1 -
Open-source Chrome extension: point at your UI, talk, and your coding agent gets the context – selector, component, styles, screenshot and errors via a local MCP server.
MIT ★ 1 -
Reproducible game capture with scenarios, replayable sessions, OBS recording, and FFmpeg derivatives.
npm MIT