Server data from the Official MCP Registry
Vantio Optics — read-only Sight Loop observe for agents. Proofs, hosts, upgrade path.
About
Vantio Optics — read-only Sight Loop observe for agents. Proofs, hosts, upgrade path.
Security Report
Valid MCP server (3 strong, 4 medium validity signals). No known CVEs in dependencies. ⚠️ Package registry links to a different repository than scanned source. Imported from the Official MCP Registry. 1 finding(s) downgraded by scanner intelligence.
16 files analyzed · 1 issue found
Security scores are indicators to help you make informed decisions, not guarantees. Always review permissions before connecting any MCP server.
Permissions Required
This plugin requests these system permissions. Most are normal for its category.
How to Install
Add this to your MCP configuration file:
{
"mcpServers": {
"io-github-vantioai-vantio-optics": {
"args": [
"-y",
"@vantio/optics-mcp"
],
"command": "npx"
}
}
}Documentation
View on GitHubFrom the project's GitHub README.
Vantio Optics (Open Core)
Free, local-first observability for AI agent egress. See what your agents call out to — without ever reading the conversation.
npm install -g @vantio/cli
vantio run node agent.js
That's the integration. Optics intercepts outbound calls to known LLM providers and reports destination, process, size, and timing — never prompts or completions.
Python: pip install vantio-agent-sdk, then wrap your agent with @shield.
Optics is Tier 1 of Vantio's architecture — Observe. Gate (Enforce) and Phantom Engine (Control) extend this to policy enforcement and host-level runtime protection on enrolled Linux systems. Full docs: vantio.ai/docs.
Supported runtimes
That's the only install step. Works on macOS, Linux, and Windows (WSL).
On Node: fetch, undici, http/https, http2, net/tls, WebSocket, and spawned curl or wget.
vantio run only injects the Node interceptor (NODE_OPTIONS --require) for node / npx / tsx / ts-node.
Python support requires vantio-agent-sdk. Follow the current Python SDK example and verify that a supported outbound event appears before relying on the coverage state. Prefixing vantio run python does not intercept by itself.
Browser paths stay outside this wrap.
What is recorded
Optics records connection and process metadata from supported wrapped agent paths. Does not block actions or retain prompts or completions.
Vantio records that a call was made, when, to which provider, and how many bytes — nothing more.
What is never recorded
- The content of your prompts
- Model completions or responses
- Any personally identifiable information
Present coverage boundaries
vantio run intercepts LLM calls by patching globalThis.fetch in the Node.js
runtime (via NODE_OPTIONS --require). This covers the vast majority of real agents
— every OpenAI SDK call, every LangChain.js call, every Vercel AI SDK call — without
any code changes.
It can be bypassed. A process can:
- Call an LLM endpoint directly through a native socket without using
fetch - Spawn a subprocess that isn't started with
vantio run - Use a language runtime other than Node without installing the Python SDK
This is intentional, not a bug. Vantio Optics surfaces your governance gap honestly.
Without the SDK, vantio run python does not intercept. Browser paths, or processes not started with vantio run / shield(), never hit Optics — and Optics does not invent a record for them.
License and telemetry
MIT License.
Anonymous, opt-out usage analytics. No prompts, completions, API keys, or emails. Opt out at any time:
export VANTIO_TELEMETRY_DISABLED=1 # or
export DO_NOT_TRACK=1
Upgrade path to Gate and Phantom Engine
Free Optics needs no account and no API key. vantio login is optional and later, for dashboard sync only, never a required step before running.
Optics is Tier 1 of Vantio's architecture — Observe. Gate (Enforce) and Phantom Engine (Control) extend this to policy enforcement and host-level runtime protection on enrolled Linux systems.
Residual risk closes with Vantio Gate, then Vantio Phantom Engine — see vantio.ai/pricing.
Reviews
No reviews yet
Be the first to review this server!
More Developer Tools MCP Servers
Git
Freeby Modelcontextprotocol · Developer Tools
Read, search, and manipulate Git repositories programmatically
Fetch
Freeby Modelcontextprotocol · Developer Tools
Web content fetching and conversion for efficient LLM usage
Toleno
Freeby Toleno · Developer Tools
Toleno Network MCP Server — Manage your Toleno mining account with Claude AI using natural language.
mcp-creator-python
Freeby mcp-marketplace · Developer Tools
Create, build, and publish Python MCP servers to PyPI — conversationally.
MCP Marketplace
Freeby mcp-marketplace · Developer Tools
Search and install MCP servers from inside your AI client.
MarkItDown
Freeby Microsoft · Content & Media
Convert files (PDF, Word, Excel, images, audio) to Markdown for LLM consumption