Resolve, search, and validate AT Protocol lexicons.
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.
Check compatibility between two versions of an ATProtocol lexicon schema. Detects breaking vs non-breaking changes and provides migration guidance. Supply either two schema JSON objects directly (from_schema + to_schema), or an NSID with two CIDs to compare stored versions (nsid + from_cid + to_cid).
Optional: The lexicon NSID. If provided with from_cid and to_cid, fetches schemas from storage.
Optional: CID of the newer version (used with nsid)
Optional: CID of the older version (used with nsid)
Optional: The newer schema JSON (used without nsid)
Optional: The older schema JSON (used without nsid)
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 lexicon-garden https://lexicon.garden/mcp
Check compatibility between two versions of an ATProtocol lexicon schema. Detects breaking vs non-breaking changes and provides migration guidance. Supply either two schema JSON objects directly (from_schema + to_schema), or an NSID with two CIDs to compare stored versions (nsid + from_cid + to_cid).
| Argument | Type | Description |
|---|---|---|
| nsid | string | Optional: The lexicon NSID. If provided with from_cid and to_cid, fetches schemas from storage. |
| to_cid | string | Optional: CID of the newer version (used with nsid) |
| from_cid | string | Optional: CID of the older version (used with nsid) |
| to_schema | object | Optional: The newer schema JSON (used without nsid) |
| from_schema | object | Optional: The older schema JSON (used without nsid) |
{
"type": "object",
"properties": {
"nsid": {
"type": "string",
"description": "Optional: The lexicon NSID. If provided with from_cid and to_cid, fetches schemas from storage."
},
"to_cid": {
"type": "string",
"description": "Optional: CID of the newer version (used with nsid)"
},
"from_cid": {
"type": "string",
"description": "Optional: CID of the older version (used with nsid)"
},
"to_schema": {
"type": "object",
"description": "Optional: The newer schema JSON (used without nsid)"
},
"from_schema": {
"type": "object",
"description": "Optional: The older schema JSON (used without nsid)"
}
}
}Get detailed information about an ATProtocol lexicon schema: its definitions, authority, CID and any published documentation. Use this to understand the structure and format of XRPC methods, records, and other lexicon types.
| Argument | Type | Description |
|---|---|---|
| lexicon* | string | The lexicon NSID to describe (e.g., 'app.bsky.feed.post', 'com.atproto.repo.createRecord', or 'app.bsky.actor.profile#viewBasic' for a specific fragment) |
| identity | string | Optional: The identity (DID like 'did:plc:...' or handle like 'alice.bsky.social') of the authority publishing this lexicon. If not provided, the authority will be resolved via DNS. |
| language | string | Optional: Preferred language code (ISO 639) for localized descriptions from garden.lexicon.documentation records. If not provided, only schema descriptions are shown. Use language codes like 'en', 'pt-BR', 'ja', etc. |
{
"type": "object",
"required": [
"lexicon"
],
"properties": {
"lexicon": {
"type": "string",
"description": "The lexicon NSID to describe (e.g., 'app.bsky.feed.post', 'com.atproto.repo.createRecord', or 'app.bsky.actor.profile#viewBasic' for a specific fragment)"
},
"identity": {
"type": "string",
"description": "Optional: The identity (DID like 'did:plc:...' or handle like 'alice.bsky.social') of the authority publishing this lexicon. If not provided, the authority will be resolved via DNS."
},
"language": {
"type": "string",
"description": "Optional: Preferred language code (ISO 639) for localized descriptions from garden.lexicon.documentation records. If not provided, only schema descriptions are shown. Use language codes like 'en', 'pt-BR', 'ja', etc."
}
}
}Discover which ATProtocol permission-set lexicon schemas cover a given set of collections and/or XRPC methods. Returns matching permission sets with their OAuth scope strings (`include:<nsid>`), useful for determining what OAuth scopes an application needs.
| Argument | Type | Description |
|---|---|---|
| methods | array | Optional: List of XRPC method NSIDs your application calls (e.g., ['app.bsky.video.uploadVideo', 'com.atproto.repo.listRecords']) |
| collections* | array | List of collection NSIDs your application reads or writes (e.g., ['app.bsky.feed.post', 'app.bsky.actor.profile', 'garden.lexicon.example']) |
{
"type": "object",
"required": [
"collections"
],
"properties": {
"methods": {
"type": "array",
"items": {
"type": "string"
},
"description": "Optional: List of XRPC method NSIDs your application calls (e.g., ['app.bsky.video.uploadVideo', 'com.atproto.repo.listRecords'])"
},
"collections": {
"type": "array",
"items": {
"type": "string"
},
"description": "List of collection NSIDs your application reads or writes (e.g., ['app.bsky.feed.post', 'app.bsky.actor.profile', 'garden.lexicon.example'])"
}
}
}Resolve an ATProtocol lexicon by NSID: find the DID named by the `_lexicon` TXT record of the NSID's authority and read that DID's com.atproto.lexicon.schema record, from Lexicon Garden's index or, when it is not indexed or refresh is set, from the network. Returns the schema, its AT-URI and CID, the authority DID and where it came from.
| Argument | Type | Description |
|---|---|---|
| nsid* | string | The NSID to resolve (e.g., 'app.bsky.feed.post') |
| refresh | boolean | Optional: read the record from the network even when it is indexed |
{
"type": "object",
"required": [
"nsid"
],
"properties": {
"nsid": {
"type": "string",
"description": "The NSID to resolve (e.g., 'app.bsky.feed.post')"
},
"refresh": {
"type": "boolean",
"description": "Optional: read the record from the network even when it is indexed"
}
}
}Find ATProtocol lexicons by what their schemas say: descriptions, definition names, field names and known values. Use it when you don't know the NSID. mode "keyword" (default) matches words; mode "semantic" ranks by meaning, so a plain description of what you need works (e.g. 'store a user's reading list'), and it falls back to keyword when unavailable. By default only lexicons published by their NSID's DNS authority are returned. Returns NSIDs with summaries and AT-URIs; follow up with describe_lexicon.
| Argument | Type | Description |
|---|---|---|
| mode | string | keyword (default) or semantic |
| limit | integer | Maximum results (default 10) |
| query* | string | What to search for: words, a phrase in quotes, -word to exclude, or (with mode semantic) a plain description. 2 to 256 characters. |
| lexicon_types | array | Only return lexicons whose main definition has one of these types |
| include_non_authoritative | boolean | Also return lexicons whose publisher is not the NSID's DNS authority (default false). Such copies may be forks or impersonations. |
{
"type": "object",
"required": [
"query"
],
"properties": {
"mode": {
"enum": [
"keyword",
"semantic"
],
"type": "string",
"description": "keyword (default) or semantic"
},
"limit": {
"type": "integer",
"maximum": 20,
"minimum": 1,
"description": "Maximum results (default 10)"
},
"query": {
"type": "string",
"description": "What to search for: words, a phrase in quotes, -word to exclude, or (with mode semantic) a plain description. 2 to 256 characters."
},
"lexicon_types": {
"type": "array",
"items": {
"enum": [
"record",
"query",
"procedure",
"subscription",
"permission-set"
],
"type": "string"
},
"description": "Only return lexicons whose main definition has one of these types"
},
"include_non_authoritative": {
"type": "boolean",
"description": "Also return lexicons whose publisher is not the NSID's DNS authority (default false). Such copies may be forks or impersonations."
}
}
}Validate a lexicon schema and optionally validate data against it. Use this to verify schema correctness and validate records, query params, or procedure inputs against their schemas.
| Argument | Type | Description |
|---|---|---|
| flags | object | Optional: Validation flags |
| input | object | Optional: Procedure input body to validate (for procedure schemas only) |
| params | object | Optional: Query parameters to validate (for query or procedure schemas) |
| record | object | Optional: Record data to validate (for record schemas only) |
| schema* | object | The lexicon schema to validate (the full lexicon document as a JSON object) |
{
"type": "object",
"required": [
"schema"
],
"properties": {
"flags": {
"type": "object",
"properties": {
"allowLegacyBlob": {
"type": "boolean",
"description": "Allow legacy blob format (cid field instead of ref)"
},
"externalSchemas": {
"type": "array",
"items": {
"type": "object"
},
"description": "External schemas to use for validation (as JSON objects)"
},
"skipExternalRefs": {
"type": "boolean",
"description": "Skip validation of external references (treat them as valid)"
},
"allowLenientDatetime": {
"type": "boolean",
"description": "Allow lenient datetime parsing"
}
},
"description": "Optional: Validation flags"
},
"input": {
"type": "object",
"description": "Optional: Procedure input body to validate (for procedure schemas only)"
},
"params": {
"type": "object",
"description": "Optional: Query parameters to validate (for query or procedure schemas)"
},
"record": {
"type": "object",
"description": "Optional: Record data to validate (for record schemas only)"
},
"schema": {
"type": "object",
"description": "The lexicon schema to validate (the full lexicon document as a JSON object)"
}
}
}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 20, 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.