Read-only MCP server for the Durable Agent Directory: search and compare agent-infrastructure tools.
Maintain this server? Claim it
Call this server's tools from your browser. Nothing to install.
Requests go through mcptry because browsers can't call most MCP servers directly. Header values are sent to the server for that request only and are never stored or logged.
Tool list from our last check. Connect live to refresh it.
The six-criteria comparison matrix for two listings, side by side.
First listing slug.
Second listing slug.
This tool isn't marked read-only, so it may change something on the server or in an account you connected.
Pick your client. Values in capitals, like YOUR_TOKEN, are yours to fill in.
Run in a terminal
claude mcp add --transport http durable-agent-directory https://agents.nexteleven.dev/api/mcp
The six-criteria comparison matrix for two listings, side by side.
| Argument | Type | Description |
|---|---|---|
| a* | string | First listing slug. |
| b* | string | Second listing slug. |
{
"type": "object",
"required": [
"a",
"b"
],
"properties": {
"a": {
"type": "string",
"description": "First listing slug."
},
"b": {
"type": "string",
"description": "Second listing slug."
}
}
}The six criteria every listing is marked against, with definitions and the support values (native = Built in, pattern = Documented pattern, none = Not claimed).
No arguments.
{
"type": "object",
"properties": {}
}Full structured record for one listing: criteria marks with notes, cost label, evidence and source links, boundaries.
| Argument | Type | Description |
|---|---|---|
| slug* | string | Listing slug, e.g. "deadbolt". |
{
"type": "object",
"required": [
"slug"
],
"properties": {
"slug": {
"type": "string",
"description": "Listing slug, e.g. \"deadbolt\"."
}
}
}Receipt status for a submission made to POST /api/lead, by its submission_id (UUID v4). Reports receipt metadata only (received, when, intent, sandbox or real) — never the inquiry contents.
| Argument | Type | Description |
|---|---|---|
| submission_id* | string | The UUID v4 used when submitting. |
{
"type": "object",
"required": [
"submission_id"
],
"properties": {
"submission_id": {
"type": "string",
"description": "The UUID v4 used when submitting."
}
}
}Free mechanical pre-check before a paid verification review: given a claims object mapping each criterion key (approvalGates, restartRecovery, evidenceTrail, killSwitch, policyControls, isolation) to a public documentation URL, reports which criteria have documentation linked and which are missing. It checks that links were supplied, never that the documentation proves the claim.
| Argument | Type | Description |
|---|---|---|
| claims* | object | Map of criterion key to public documentation URL. |
{
"type": "object",
"required": [
"claims"
],
"properties": {
"claims": {
"type": "object",
"properties": {
"isolation": {
"type": "string"
},
"killSwitch": {
"type": "string"
},
"approvalGates": {
"type": "string"
},
"evidenceTrail": {
"type": "string"
},
"policyControls": {
"type": "string"
},
"restartRecovery": {
"type": "string"
}
},
"description": "Map of criterion key to public documentation URL."
}
}
}Keyword search over the directory catalog (names, descriptions, tags, criteria marks). Returns matching listings with scores and what matched. This is keyword matching, not semantic search.
| Argument | Type | Description |
|---|---|---|
| query* | string | Question-shaped or keyword query, e.g. "self-hosted tool with approval gates". |
{
"type": "object",
"required": [
"query"
],
"properties": {
"query": {
"type": "string",
"description": "Question-shaped or keyword query, e.g. \"self-hosted tool with approval gates\"."
}
}
}We read every tool's name, description and inputs the way the model will, looking for instructions hidden from you. A clean result means no warning signs, not a guarantee. Scanned Nov 14, 2026.
No readOnlyHint / destructiveHint annotations, so clients can't tell a read from a write without asking.
Fix: Add annotations to each tool, e.g. { readOnlyHint: true } for tools that only read.