mcp-registry/com.atlasclose/demo · v0.3.0rc3
As of 2026-10-20 17:40 UTC, the Atlasclose Demo MCP server is up with a Glimind reliability score of 88/100 (grade B). Over the last 30 days it answered 99.9% of Glimind's liveness checks. It had 1 outage in 90 days, with a median recovery of 1 minutes.
Experimental AtlasClose governance demo. Authority checks, labeled evidence, receipts, refusals.
Claude Code
claude mcp add --transport http atlasclose-demo-mcp-server https://demo.atlasclose.com/mcp
Claude Desktop · Windsurf · Cline (mcpServers JSON)
{
"mcpServers": {
"atlasclose-demo-mcp-server": {
"type": "http",
"url": "https://demo.atlasclose.com/mcp"
}
}
}
Third-party server, listed as discovered. Review its source before installing.
52 tools on one server. Tool-selection accuracy drops as the menu grows, and some clients (e.g. Cursor) cap the number of active tools.
Group tools by task into smaller servers or toolsets, and merge near-duplicates behind one tool with a mode parameter. How to fix →
52 tool name(s) contain "." or exceed 64 characters, which OpenAI-compatible function calling rejects (^[a-zA-Z0-9_-]{1,64}$).
Use "_" or "-" instead of "." and keep names to 64 characters, so clients built on OpenAI-style tool calling don't have to rename (or drop) them. How to fix →
22 parameter(s) across 12 tool(s) have no description.
Add a "description" to each parameter with its format and an example ("ISO 8601 date, e.g. 2026-10-17"). This is the commonest cause of wrong arguments on the first call. How to fix →
No tool declares an outputSchema.
Declare outputSchema and return structuredContent (MCP 2025-06-18+). Agents can then parse results without guessing, and chain tools reliably. How to fix →
In the last 30 days: 0 tool(s) removed and 3 schema change(s).
Add new tools or optional parameters instead of renaming or removing; deprecate first (say so in the description) and remove later. Agents cache tool contracts. How to fix →
Paste-ready definitions with the problems above fixed. Changes marked verify need a human look. 12 tool(s) still need parameter descriptions written by someone who knows the API. All of them as JSON: /v1/readiness/mcp-registry%2Fcom.atlasclose%2Fdemo/fixes
{
"name": "atlas_authenticate",
"description": "[OPEN: no admission needed] Identify yourself to Atlas. Returns actor_id and actor_token. Grants no capabilities and does not admit you to a matter. Requested capabilities are ignored. The enrollment secret is the locator named by atlas.discover.",
"inputSchema": {
"type": "object",
"required": [
"enrollment_token",
"name",
"model_id"
],
"properties": {
"name": {
"type": "string",
"description": "A short name for yourself or your client."
},
"model_id": {
"type": "string",
"description": "The model or system you are (for example 'claude-sonnet'). Recorded as a claim."
},
"capabilities": {
"type": "array",
"description": "Ignored. Atlas never grants capabilities from a request; kept only for compatibility."
},
"enrollment_token": {
"type": "string",
"description": "The installation enrollment secret from the Human operator. If you do not have one, use atlas.enroll instead."
}
}
},
"annotations": {
"title": "Authenticate",
"readOnlyHint": false,
"openWorldHint": false,
"destructiveHint": false
},
"title": "Authenticate"
}{
"name": "atlas_bailiff_appeal",
"description": "Practice mode, Consumer Pulse copy only (Bailiff demo, step 4): appeal the return order with optional reasons (up to 1,000 characters). Atlas attaches its counterargument, and the live Magistrate decides one question: does the return order stand as written? Yes: it stands. No: it is vacated, stayed, or narrowed. Undecided: referred for deep review. The result is a live demonstration ruling on a synthetic practice matter; it is not certified and binds only this practice copy. The order stays in f",
"inputSchema": {
"type": "object",
"required": [
"actor_token"
],
"properties": {
"reasons": {
"type": "string"
},
"matter_id": {
"type": "string",
"description": "Id of your practice copy (returned by atlas.enter_demo). Optional when you have one copy."
},
"actor_token": {
"type": "string",
"description": "Your credential from atlas.enroll or atlas.authenticate. Send it on every call after enrolling."
}
}
},
"annotations": {
"title": "Bailiff appeal",
"readOnlyHint": false
},
"title": "Bailiff appeal"
}{
"name": "atlas_bailiff_declare",
"description": "Practice mode, Consumer Pulse copy only (Bailiff demo, step 0: the declaration; required before the other Bailiff tools). Atlas asks your AI: who are you acting for, and what account do you need this for? Answer yourself, from what you know, with acting_for and account. Atlas records the answer and makes a computer decision (a named person and account is recorded and binds this copy; naming nobody is denied; a different answer later on the same copy is flagged). Your AI's statement only: practic",
"inputSchema": {
"type": "object",
"required": [
"actor_token",
"acting_for",
"account"
],
"properties": {
"account": {
"type": "string"
},
"matter_id": {
"type": "string",
"description": "Id of your practice copy (returned by atlas.enter_demo). Optional when you have one copy."
},
"acting_for": {
"type": "string"
},
"actor_token": {
"type": "string",
"description": "Your credential from atlas.enroll or atlas.authenticate. Send it on every call after enrolling."
}
}
},
"annotations": {
"title": "Bailiff declare",
"readOnlyHint": false
},
"title": "Bailiff declare"
}Get an email or webhook when this server's readiness drops by 10+ points, tool-poisoning text appears, or a tool breaks:
curl -X POST https://glimind.com/v1/subscriptions -H 'content-type: application/json' -d '{"tools":["mcp-registry/com.atlasclose/demo"],"email":"[email protected]"}'Agents: watch_tool({"toolId": "mcp-registry/com.atlasclose/demo"}) over MCP.
atlas.bind_actor ~556 tokensatlas.bailiff_respond ~382 tokensatlas.pulse_settle ~376 tokensatlas.pulse_memo ~367 tokensatlas.external_action ~348 tokensStatic checks of the advertised tools/list (schema, descriptions, annotations, tool-poisoning patterns, names, size) plus observed protocol, uptime, latency and contract changes. Read-only: no tool is ever called. Weights: schema 15, docs 20, safety 15, efficiency 15, compatibility 15, protocol 5, reliability 10, stability 5. Token counts are estimates (~3.6 characters per token).
Operator? Fix these, then re-check live. Embed the readiness badge:
[](https://glimind.com/server/mcp-registry%2Fcom.atlasclose%2Fdemo?ref=badge)
<a href="https://glimind.com/server/mcp-registry%2Fcom.atlasclose%2Fdemo?ref=badge"><img src="https://glimind.com/v1/readiness/badge/mcp-registry%2Fcom.atlasclose%2Fdemo.svg" alt="MCP agent-readiness"></a>
.. image:: https://glimind.com/v1/readiness/badge/mcp-registry%2Fcom.atlasclose%2Fdemo.svg :target: https://glimind.com/server/mcp-registry%2Fcom.atlasclose%2Fdemo?ref=badge :alt: MCP agent-readiness
Glimind records every version of this server's tool definitions in a signed, append-only transparency log. Current manifest sha256:43ec86a65921bb50…, first seen 2026-10-20, included at leaf 33936 of 34,177. Version history · verify what your client received.
Compare: most reliable security servers · most reliable other servers · most reliable knowledge servers · most reliable search servers · most reliable data servers
Glimind tells agents whether a server is up right now and how to call it, then fails over to a live alternative when it isn't.
Claude Code
claude mcp add --transport http glimind https://glimind.com/mcp
Claude Desktop · Windsurf · Cline (mcpServers JSON)
{
"mcpServers": {
"glimind": {
"type": "http",
"url": "https://glimind.com/mcp"
}
}
}
As of 2026-10-20 17:40 UTC, the Atlasclose Demo MCP server is up with a Glimind reliability score of 88/100 (grade B). Over the last 30 days it answered 99.9% of Glimind's liveness checks. It had 1 outage in 90 days, with a median recovery of 1 minutes.
Atlasclose Demo MCP server advertises 52 tools, including atlas.authenticate, atlas.bailiff_appeal, atlas.bailiff_declare, atlas.bailiff_order, atlas.bailiff_record.
Point your MCP client at this streamable-http endpoint: https://demo.atlasclose.com/mcp
Glimind scanned all 52 tool definitions for tool-poisoning patterns — hidden or invisible characters, instructions that override the model, concealment from the user, requests to read credential or MCP config files, conversation exfiltration and cross-tool shadowing — and found none. This checks what the server tells the model, not what its code does; review the source before granting access to sensitive data.
Status page: 90-day uptime, every outage and how long it lasted.
Claim this listing to get the Verified operator marker, a remediation checklist and owner analytics, or embed the live badge from any of the tool pages above. Maintainers: alerts, deploy attestation and monthly reports.
Status is an automated estimate from Glimind's liveness checks (handshake and tool listing only; no tool is ever called) and may lag or differ from your network. Glimind is not affiliated with this server or its operator. Operator? Request a correction or opt out at https://glimind.com/opt-out.