Openeye MCP Server

Procedure verification for AR/XR with memory, visual sessions, and DPO trajectory export.

Remote serverstreamable-httpPython

What is the Openeye MCP server?

Procedure verification for AR/XR with memory, visual sessions, and DPO trajectory export. Exposed over MCP by the openeye mcp server, that capability becomes something an assistant can invoke while it works, not something you go and do afterwards.

What it actually does

every AR headset maker — HoloLens, Snap Spectacles, Apple, Android — ships a device that can see. none of them ship a brain. every developer building on these platforms has to figure out the intelligence layer themselves, from scratch, every time, for every device. the work doesn't compound. what one team learns on HoloLens doesn't help the team building on WebXR. every deployment is a silo.

Its toolset

Everything the assistant can do here goes through one of these:

  • capability — OpenEye
  • search_memory — FTS5 search across all past agent sessions
  • search_frames — FTS5 search across all past frame descriptions
  • recall_skill — retrieve relevant procedural skills for the current task
  • write_skill — persist a new skill doc after completing a complex task
  • start_visual_session — begin a tracked AR/XR session
  • end_visual_session — close a visual session
  • log_frame — record a frame's scene description
  • verify_step — record a step result — pass / fail / uncertain
  • provider — env var
  • OPENEYE_HOME — ~/.openeye
  • OPENEYE_PORT — 7770

Adding it to your client

Being a remote server, there is no local install. You register the endpoint with your client, authorise it once, and the tools appear.

Configuration

You will need 8 environment variables: ANTHROPIC_API_KEY, OPENEYE_CONTEXT_OPTIN, OPENEYE_CONTEXT_API_KEY, GROQ_API_KEY, OPENEYE_BIND_HOST, OPENEYE_SIDECAR_TOKEN, OPENEYE_CLOUD_URL, OPENEYE_CLOUD_KEY. The server will not start without them, which is usually why the tools fail to appear on a first run. Keep credentials in your client's env block or a secrets manager rather than in a file you might commit.

When to reach for it

Plenty of planning and project tracking servers cover similar ground. The differences that matter in practice are scope of access and how much setup stands between you and a working tool call. Openeye's toolset — capability, search_memory, search_frames and 11 more — is a fair guide to whether it matches your workflow. It is maintained by dumbspacecookie; worth a glance at recent repository activity before you build anything load-bearing on it.

SyncDev reviews every entry in this directory against the project's own documentation before publishing, and revisits them as servers change.

Caveats

  • Your data travels to the provider's service, so the usual questions apply about what you send and what they retain.
  • With 14 tools registered it takes up a noticeable share of the context window; turn it off in projects that never touch Openeye.
  • Missing credentials fail quietly in some clients — if no tools show up, check the environment block first.
  • MCP clients confirm each tool call by default. Leave that on until you have watched what the openeye mcp server does with a few real requests.

Available tools

ToolWhat it does
capabilityOpenEye
search_memoryFTS5 search across all past agent sessions
search_framesFTS5 search across all past frame descriptions
recall_skillretrieve relevant procedural skills for the current task
write_skillpersist a new skill doc after completing a complex task
start_visual_sessionbegin a tracked AR/XR session
end_visual_sessionclose a visual session
log_framerecord a frame's scene description
verify_steprecord a step result — pass / fail / uncertain
providerenv var
OPENEYE_HOME~/.openeye
OPENEYE_PORT7770
OPENEYE_BIND_HOST127.0.0.1
OPENEYE_SIDECAR_TOKEN

How to install the Openeye MCP server

{
  "mcpServers": {
    "openeye": {
      "command": "npx",
      "args": ["-y", "@dumbspacecookie/openeye"],
      "env": {
        "ANTHROPIC_API_KEY": "your-value",
        "OPENEYE_CONTEXT_OPTIN": "your-value",
        "OPENEYE_CONTEXT_API_KEY": "your-value",
        "GROQ_API_KEY": "your-value",
        "OPENEYE_BIND_HOST": "your-value",
        "OPENEYE_SIDECAR_TOKEN": "your-value",
        "OPENEYE_CLOUD_URL": "your-value",
        "OPENEYE_CLOUD_KEY": "your-value"
      }
    }
  }
}

Add to claude_desktop_config.json, then restart Claude Desktop.

Configuration

VariableDescriptionRequired
ANTHROPIC_API_KEYCredential the server authenticates with.Yes
OPENEYE_CONTEXT_OPTINConfiguration value read at startup.Optional
OPENEYE_CONTEXT_API_KEYCredential the server authenticates with.Yes
GROQ_API_KEYCredential the server authenticates with.Yes
OPENEYE_BIND_HOSTEndpoint or connection string the server talks to.Optional
OPENEYE_SIDECAR_TOKENCredential the server authenticates with.Yes
OPENEYE_CLOUD_URLEndpoint or connection string the server talks to.Yes
OPENEYE_CLOUD_KEYCredential the server authenticates with.Yes

Example prompts to try

  • Use Openeye to capability.
  • Use Openeye to search memory.
  • Use Openeye to search frames.

Frequently asked questions

It connects Openeye to MCP-compatible AI assistants such as Claude and Cursor, exposing 14 tools (capability, search_memory, search_frames, and more) that the assistant can call on your behalf. Instead of copying data back and forth by hand, the assistant works with Openeye directly.