Agenvoy
Self-hosted AI agent that auto-builds sandboxed tools and shares them via MCP
If you want a self-hosted agent that builds and shares tools across MCP clients, Agenvoy delivers. Auto-tool-generation and multi-model dispatch are genuinely useful. But the CLI and setup overhead mean non-technical users should look elsewhere. Pick it for ownership and control, not zero-config convenience.
Verified 4d ago · liveness 69/100 · cite: rightaichoice.com/tools/agenvoy
- Developers who want a self-hosted, always-on AI assistant
- Power users needing a model-agnostic agent with tool-building and scheduling
- Teams wanting a shared sandboxed tool library across MCP-compatible agents
- Privacy-conscious users who need offline capability without cloud dependencies
- Users who prefer a purely hosted, no-install AI assistant
- Non-technical users wanting a GUI-first, no-code experience
- Teams needing enterprise-grade support SLAs or managed hosting
We scan live Reddit threads, YouTube comments, X posts, G2 reviews and other communities — and hand you an honest verdict in under a minute.
- Honest verdict, not marketing
- Real pros & cons from real users
- Attributed quotes with receipts
3 free scans · no card needed
Skip Agenvoy if you're not comfortable running a command-line daemon on your own machine, if you need a hosted, zero-install assistant, or if you rely on enterprise support and managed infrastructure — this tool is built for developers who own their stack.
You need to run a persistent Go daemon on your own hardware — electricity and maintenance are on you.
Agenvoy is free and open-source (Apache 2.0), which undercuts most hosted AI assistants like ChatGPT Pro ($20/mo) or Claude Pro ($20/mo). But you pay indirectly through your own compute and LLM API costs. For developers who already pay for APIs, this can be cheaper at scale, but for non-technical users who want a turnkey solution, free hosted tiers might be more cost-effective.
In short
Agenvoy — Self-hosted AI agent that auto-builds sandboxed tools and shares them via MCP. Best for Developers who want a self-hosted, always-on AI assistant, Power users needing a model-agnostic agent with tool-building and scheduling, Teams wanting a shared sandboxed tool library across MCP-compatible agents. Free to use.
What people actually say about Agenvoy — is it worth it?
We ran a structured research pass across product reviews, community discussions, and post-purchase forum threads to surface the patterns vendors won't publish themselves. Below: the recurring strengths, the hidden costs people mention most, and the cohort that consistently regrets adopting this tool.
13 mentions across 3 sources (Hacker News, GitHub, Lemmy) · researched Aug 14, 2026.
- +One-command curl install makes deployment quick and painless.
- +Auto-generates and tests scripts, removing manual tool coding.
- +Dispatches tasks across 12+ LLM providers intelligently.
- +OS-native sandboxing (bubblewrap, sandbox-exec) provides strong safety.
- +Three-tier memory system supports long-term context and recall.
- −history.json grows unbounded, causing OOM after hours of use.
- −Updating versions breaks tool access in existing sessions.
- −Gemini function calling fails with HTTP 400 on multi-tool turns.
- −Autonomous execution of long plans (40+ steps) errors out.
- −TUI displays wrong version number, confusing users.
- • No monetary hidden costs, but time is needed for setup and debugging.
Viability Score
How well maintained and how widely used is Agenvoy? Built from what the vendor actually publishes (docs, changelog, tutorials, integrations, pricing), whether the site is live, and how much real users discuss it. How we calculate this
Last calculated: September 2026
How we score →Key Features
- Auto tool generation - writes, tests, and registers scripts
- MCP server mode - expose tools to Claude Code, Codex, etc.
- MCP client mode - call external MCP services
- Multi-model dispatch across 10+ providers
- Cron scheduler with skill binding and hot-reload
- 3-tier memory: context window, vector search, SQLite archive
- Cross-session error memory with 90-day TTL
- Native document RAG via KuraDB child process
- Loadable markdown skill packs with YAML frontmatter
- Self-improvement - rewrites failed skills, git-commits, retries
- OS-native sandboxed execution (bubblewrap, sandbox-exec)
- Web dashboard (local browser connection)
- TUI with slash commands and session switching
- Telegram integration with OTP and button confirms
- Discord integration with select menus and modals
About Agenvoy
Agenvoy is a self-hosted AI agent harness that runs as a single Go daemon on your machine. It turns plain-English requests into working tools: when no existing tool fits, the agent finds an API, writes a script, tests it in a sandbox, and registers it for reuse. You interact through Telegram, Discord, a TUI, or the new web dashboard at web.agenvoy.com, which connects to your local agent without data leaving your device. Under the hood, Agenvoy routes tasks across 10 LLM providers—Anthropic Claude, OpenAI GPT and Codex, Gemini, GitHub Copilot, NVIDIA NIM, xAI Grok, DeepSeek, Mistral, OpenRouter, and local models via Ollama, LM Studio, vLLM. A dispatcher automatically picks the best model per task. Every command and script executes inside an OS-native sandbox (bubblewrap on Linux, sandbox-exec on macOS) to keep your system safe. Core capabilities include a three-tier memory system (context window, semantic vector search, SQLite full-text archive) with cross-session error memory that persists fixes and their success. A cron scheduler handles jobs and one-shot tasks with skill binding, hot-reload, and auto-disable on failure. The skill system lets you load markdown instruction packs with YAML frontmatter, triggered by slash commands or natural language. When a tool fails, the agent rewrites it, git-commits the fix, and retries, with rollback via file history. Agenvoy also includes MCP server and client modes. Server mode exposes all sandboxed tools to Claude Code, Codex, OpenCode, Cursor, and any MCP-compatible agent; client mode connects external MCP services via stdio or HTTP/SSE. The tool library ships with 25 built-in tools, covering files, web, execution, coordination, and extensibility via script_*, api_*, ext_*, and mcp__* namespaces. It's Apache 2.0 licensed, one binary, installable in 60 seconds. Compared to hosted assistants, Agenvoy gives you full control and offline capability at the cost of setup effort.
Behind the Verdict
Agenvoy is for developers who want an always-on, self-hosted agent without vendor lock-in. It's a single binary that installs in 60 seconds and runs entirely on your machine, so your data never leaves your device. The auto-tool-generation is the standout: ask for anything, and it writes and tests a script, registers it, and reuses it — that's a feature few hosted assistants offer. Where it shines: if you live in Claude Code, Cursor, or Codex, Agenvoy's MCP server mode hands your entire sandboxed tool library to those clients, and they can build new tools that everyone shares. The multi-model dispatcher across 10 providers plus local models means you can optimize for cost and quality per task. The cron scheduler with Telegram push is a nice touch for daily automation. Where it bites: setup is not zero-config. You need to install the daemon, add provider keys, and be comfortable with a TUI or a web dashboard. There's no polished mobile app or managed hosting, so if you want a plug-and-play assistant, look elsewhere. The sandboxing is OS-native, so you need a compatible Linux or macOS environment. Compared to hosted assistants like ChatGPT or Claude, Agenvoy trades convenience for control and privacy. If you're a tinkerer who enjoys scripting and extending an agent via skills and MCP, this is a solid pick — but if you're non-technical, it's a poor fit.
Researching Agenvoy? Get your full AI stack in 60 seconds.
Free, no signup — tell us your goal and get tools matched to your budget & existing stack.
Real-world workflow fit
Concrete scenarios for the personas Agenvoy actually fits — and what changes day-one when you adopt it.
Wants daily stock price reports on Telegram at 8am.
Outcome: Asks the agent in plain English; it auto-builds a Yahoo Finance tool, schedules a cron job, and pushes results to Telegram every morning.
Uses Claude Code and Codex and wants one shared tool library.
Outcome: Adds one config line to expose all Agenvoy tools via MCP server; both agents can access and auto-build tools that all agents share.
Needs to search personal knowledge base without cloud exposure.
Outcome: Enables KuraDB RAG, asks questions in the TUI or Telegram, gets keyword + semantic search results entirely on-device.
Use Cases
- Automate daily stock price reports via Telegram using scheduled skills and Yahoo Finance tool.
- Build a custom weather tool on-the-fly by asking the agent, which auto-creates and registers the script.
- Share your entire tool library with Claude Code or Codex via a single MCP config line.
- Use RAG to search local documents with KuraDB while chatting in the TUI or Telegram.
- Set up a cron job that fetches live data, processes it, and pushes results to Discord with output push-back.
- Create a skill pack for code review that invokes sub-agents and external agents for parallel review.
- Call external MCP services like GitHub via client mode and auto-register their tools.
Models Under the Hood
as of 2026-08-30
Limitations
- Agenvoy is a self-hosted AI agent harness that requires a Go daemon running on your local machine, with a web dashboard available only when the daemon is running and connected to your local agent.
- Sandboxing relies on OS-native utilities (bubblewrap on Linux, sandbox-exec on macOS), and multi-model dispatch quality depends on provider availability and configuration.
- The project is open-source under Apache 2.0 and is built in Taiwan, with no hosted option mentioned.
as of 2026-08-21
Verification history
We have re-verified Agenvoy 7 times since . Each pass re-reads the vendor's own pages and re-checks every listed field against that evidence; passes where nothing had changed are marked as such.
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-checked, vendor evidence unchanged
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
Showing the 6 most recent of 7 verification passes.
Free to cite with attribution — this page re-verifies continuously.
Where the pricing makes sense
The company stage and team size where Agenvoy's pricing actually pencils out — and where peers do it cheaper.
Agenvoy is free and open-source (Apache 2.0), which undercuts most hosted AI assistants like ChatGPT Pro ($20/mo) or Claude Pro ($20/mo). But you pay indirectly through your own compute and LLM API costs. For developers who already pay for APIs, this can be cheaper at scale, but for non-technical users who want a turnkey solution, free hosted tiers might be more cost-effective.
Setup time & first value
How long it actually takes to get something useful out of Agenvoy — broken out by persona, not the marketing-page minute.
For a developer comfortable with the terminal: under 10 minutes to install via curl script, run the daemon, and start chatting in the TUI. Telegram/Discord setup adds another 10-15 minutes. Non-technical users may take 30+ minutes to configure sandbox and providers. First value (e.g., a simple weather query) can be had within minutes of install.
Switching to or from Agenvoy
How to bring data in from common predecessors and how to get it back out — written for the switcher, not the buyer.
- →From a hosted assistant like ChatGPT: you'll lose zero-config convenience but gain local control and tool-building via MCP.
- →From a similar self-hosted agent like AutoGPT: Agenvoy offers a more streamlined sandboxing and MCP integration out of the box.
- ↗To a hosted platform (e.g., ChatGPT, Claude) if you need zero-maintenance and don't mind cloud processing.
- ↗To a different self-hosted agent if you prefer a different architecture or need features not in Agenvoy.
Integrations
Resources & Guides
- Quickstartagenvoy.com
Quickstart · Agenvoy
Get up and running fast from agenvoy.com
- Documentationagenvoy.com
Tui Guide · Agenvoy
Full product docs from agenvoy.com
- Documentationagenvoy.com
Rest Api · Agenvoy
Full product docs from agenvoy.com
- Documentationagenvoy.com
Memory System · Agenvoy
Full product docs from agenvoy.com
- Documentationagenvoy.com
Mcp Server · Agenvoy
Full product docs from agenvoy.com
- Documentationagenvoy.com
Mcp Client · Agenvoy
Full product docs from agenvoy.com
Tutorials & Learning
Official links
Tools that pair well with Agenvoy
Common stack mates teams adopt alongside Agenvoy, with the specific reason each pairing earns its keep.
Featured Head-to-Head Comparisons
Agenvoy vs Spider Cloud
If you need a self-hosted AI assistant that can autonomously write tools, schedule tasks, and interface with multiple LLMs, Agenvoy is unmatched. If your focus is extracting large-scale web data reliably for AI agents or RAG, Spider Cloud’s blazing fast Rust engine and new Browser AI commands make it the better pick. They’re complementary — use both together for maximum automation.
Agenvoy vs Presto Voice
For most developers seeking a free, self-hosted AI assistant with tool-building and MCP support, Agenvoy wins hands-down. Presto Voice is a specialized enterprise platform for QSR chains, not for individual users. Choose based on your domain: developer automation vs. drive-thru automation.
Agenvoy vs Temporal Ai
Choose Agenvoy if you're a developer wanting a self-hosted, always-on AI assistant that auto-creates tools and routes tasks across models. Choose Temporal if your priority is building fault-tolerant, durable workflows for AI agents and microservices that survive failures without losing progress. Agenvoy is a single-daemon tool-building agent; Temporal is a full workflow orchestration platform.
Alternatives to Agenvoy
View all
QuantDinger
Self-hosted AI quant trading platform with MCP integration for AI agents.
Frequently Asked Questions
Used Agenvoy? Help shape our editorial sentiment research.