feat(mcp): add Airtable read-only adapter

.github/ISSUE_TEMPLATE/config.yml

@@ -5,7 +5,7 @@ contact_links:
     about: For anything that isn't a clear bug or feature request — please use Discussions so the answer helps everyone else too.
   - name: 📚 Documentation & Guides
     url: https://anythingmcp.com
-    about: Setup guides for all 189 adapters and every AI client, in seven languages.
+    about: Setup guides for all 190 adapters and every AI client, in seven languages.
   - name: 🔐 Security vulnerabilities
     url: https://github.com/HelpCode-ai/anythingmcp/blob/main/SECURITY.md
     about: Please do NOT open a public issue. Follow the security disclosure policy.

.github/workflows/docker-publish.yml

@@ -65,7 +65,7 @@ jobs:
           readme-filepath: ./README.md
           # Docker Hub caps this at 100 characters and shows it in search
           # results; the repo's was left at the truncated default.
-          short-description: "189 connectors + any REST/SOAP/GraphQL/SQL as MCP tools for Claude & ChatGPT. Self-hosted, AGPL."
+          short-description: "190 connectors + any REST/SOAP/GraphQL/SQL as MCP tools for Claude & ChatGPT. Self-hosted, AGPL."
 
       - name: Set Docker Hub categories
         env:

CITATION.cff

@@ -13,7 +13,7 @@ abstract: >
   server that exposes REST, SOAP/WSDL and GraphQL APIs, SQL/NoSQL databases
   and other MCP servers as tools for AI clients such as Claude, ChatGPT,
   Google Gemini, GitHub Copilot and Cursor, without writing code. It ships
-  189 pre-built adapters (Deutsche Bahn, DATEV, weclapp, DHL, Shopware,
+  190 pre-built adapters (Deutsche Bahn, DATEV, weclapp, DHL, Shopware,
   Personio, Handelsregister, etc.), a per-workspace knowledge graph served
   over MCP, per-tool response mapping, OAuth2/RBAC/SSO/SCIM governance, an
   audit log and a visual tool editor.

README.md

@@ -1,12 +1,12 @@
 <p align="center">
-  <img src="docs/assets/banner.png" alt="AnythingMCP — 189 connectors, 20 of them with no API key. Your REST, SOAP/WSDL, GraphQL, SQL and MCP systems become tools for Claude, ChatGPT, Copilot and Gemini." width="100%" />
+  <img src="docs/assets/banner.png" alt="AnythingMCP — 190 connectors, 20 of them with no API key. Your REST, SOAP/WSDL, GraphQL, SQL and MCP systems become tools for Claude, ChatGPT, Copilot and Gemini." width="100%" />
 </p>
 
 <h1 align="center">AnythingMCP</h1>
 
 <p align="center">
   <strong>Give Claude, ChatGPT and Copilot safe access to the software your company already runs.</strong><br/>
-  189 ready adapters, any REST/SOAP/GraphQL/SQL system without code, on your own infrastructure — and it learns how your systems connect.
+  190 ready adapters, any REST/SOAP/GraphQL/SQL system without code, on your own infrastructure — and it learns how your systems connect.
 </p>
 
 <p align="center">
@@ -36,7 +36,7 @@ docker compose up -d   # → http://localhost:3000
 
 Three words appear throughout and mean three different things:
 
-- an **adapter** is one of the 189 JSON definitions that ship in this repo — DATEV, weclapp, DHL, Deutsche Bahn, Shopware, Personio, Handelsregister and the rest. 20 of them need no API key at all; the others ask for your credentials at import.
+- an **adapter** is one of the 190 JSON definitions that ship in this repo — DATEV, weclapp, DHL, Deutsche Bahn, Shopware, Personio, Handelsregister and the rest. 20 of them need no API key at all; the others ask for your credentials at import.
 - a **connector** is an adapter, or your own OpenAPI spec / Postman collection / WSDL / GraphQL endpoint / database, once you have configured it in your workspace. Anything you can point at, in minutes, without writing an MCP server.
 - an **MCP server** is the URL you hand to Claude. It exposes the connectors you assign to it, and nothing else.
 
@@ -260,7 +260,7 @@ AI clients speak MCP, but your systems speak REST, SOAP, GraphQL and SQL. Writin
 
 ## The adapter catalog
 
-189 adapters, exposing 1,800+ tools. **20 need no API key**; the rest ask for your credentials at import and the tools are available immediately. Every one has a setup guide on [anythingmcp.com/guides](https://anythingmcp.com/guides), in seven languages.
+190 adapters, exposing 1,800+ tools. **20 need no API key**; the rest ask for your credentials at import and the tools are available immediately. Every one has a setup guide on [anythingmcp.com/guides](https://anythingmcp.com/guides), in seven languages.
 
 | Category | Examples |
 |---|---|

glama.json

@@ -3,5 +3,5 @@
   "maintainers": [
     "keysersoft"
   ],
-  "description": "Give Claude, ChatGPT and Copilot safe access to the software your company already runs. 189 pre-built adapters (Deutsche Bahn, DATEV, weclapp, DHL, Shopware, Personio\u2026) plus any REST, SOAP/WSDL, GraphQL or SQL system as MCP tools, no code. Self-hosted, knowledge graph, per-tool response mapping, OAuth2/RBAC/SSO/audit. Open source under AGPL-3.0."
+  "description": "Give Claude, ChatGPT and Copilot safe access to the software your company already runs. 190 pre-built adapters (Deutsche Bahn, DATEV, weclapp, DHL, Shopware, Personio\u2026) plus any REST, SOAP/WSDL, GraphQL or SQL system as MCP tools, no code. Self-hosted, knowledge graph, per-tool response mapping, OAuth2/RBAC/SSO/audit. Open source under AGPL-3.0."
 }

package.json

@@ -1,7 +1,7 @@
 {
   "name": "anythingmcp",
   "version": "0.9.0",
-  "description": "Give Claude, ChatGPT and Copilot safe access to the software your company already runs. 189 pre-built adapters plus any REST, SOAP/WSDL, GraphQL or SQL system as MCP tools, no code. Self-hosted, open source (AGPL-3.0).",
+  "description": "Give Claude, ChatGPT and Copilot safe access to the software your company already runs. 190 pre-built adapters plus any REST, SOAP/WSDL, GraphQL or SQL system as MCP tools, no code. Self-hosted, open source (AGPL-3.0).",
   "private": true,
   "license": "AGPL-3.0-only",
   "engines": {

packages/backend/src/adapters/catalog.ts

@@ -41,6 +41,7 @@ import * as activecampaign from './intl/activecampaign.json';
 import * as acuityScheduling from './intl/acuity-scheduling.json';
 import * as adyen from './intl/adyen.json';
 import * as agilecrm from './intl/agilecrm.json';
+import * as airtable from './intl/airtable.json';
 import * as amadeus from './intl/amadeus.json';
 import * as amazonSeller from './intl/amazon-seller.json';
 import * as apiFootball from './intl/api-football.json';
@@ -346,6 +347,7 @@ const RAW_ADAPTERS: AdapterDefinition[] = [
   acuityScheduling as unknown as AdapterDefinition,
   adyen as unknown as AdapterDefinition,
   agilecrm as unknown as AdapterDefinition,
+  airtable as unknown as AdapterDefinition,
   amadeus as unknown as AdapterDefinition,
   amazonSeller as unknown as AdapterDefinition,
   apiFootball as unknown as AdapterDefinition,

packages/backend/src/adapters/intl/airtable.json

@@ -0,0 +1,125 @@
+{
+  "slug": "airtable",
+  "name": "Airtable",
+  "description": "Read Airtable bases, table schemas, and records from any AI agent. 4 read-only tools, Personal Access Token auth.",
+  "instructions": "This connector uses the Airtable Web API v0 and is read-only.\n\n**Setup**:\n1. Create a Personal Access Token at https://airtable.com/create/tokens.\n2. Grant only the `data.records:read` and `schema.bases:read` scopes.\n3. Restrict the token's access to only the bases this connector must read.\n4. Set `AIRTABLE_PERSONAL_ACCESS_TOKEN`.\n\n**Authentication**: `Authorization: Bearer ${AIRTABLE_PERSONAL_ACCESS_TOKEN}`. Use a scoped Personal Access Token, not a legacy API key.\n\n**Discovery order**: call `airtable_list_bases`, then `airtable_list_tables` with a base ID to find table IDs, names, fields, and views. Use a table ID when possible because names can change.\n\n**Records**: `airtable_list_records` returns one page and may include an `offset`. Pass that `offset` to the next call until it is absent. `pageSize` is 1-100; `maxRecords` limits the total records Airtable returns.\n\n**Filtering**: `filterByFormula` uses Airtable formula syntax. For example, `{Status}=\"Active\"` returns records whose Status field is Active. A view can apply saved filters and sorting; combine it with a formula only when both restrictions are intended.\n\n**Out of scope here**: creating, updating, deleting, or otherwise modifying bases, tables, fields, views, or records.",
+  "region": "intl",
+  "category": "productivity",
+  "icon": "airtable",
+  "docsUrl": "https://airtable.com/developers/web/api/introduction",
+  "requiredEnvVars": ["AIRTABLE_PERSONAL_ACCESS_TOKEN"],
+  "connector": {
+    "name": "Airtable Web API v0",
+    "type": "REST",
+    "baseUrl": "https://api.airtable.com/v0",
+    "authType": "BEARER_TOKEN",
+    "authConfig": {
+      "token": "{{AIRTABLE_PERSONAL_ACCESS_TOKEN}}"
+    },
+    "healthcheckPath": "/meta/bases"
+  },
+  "tools": [
+    {
+      "name": "airtable_list_bases",
+      "description": "List bases accessible to the Personal Access Token. Use this first to discover a base ID, for example `appABC123DEF456`.",
+      "parameters": { "type": "object", "properties": {} },
+      "endpointMapping": {
+        "method": "GET",
+        "path": "/meta/bases"
+      }
+    },
+    {
+      "name": "airtable_list_tables",
+      "description": "List a base's table schemas, including table IDs, names, fields, and views. Use this to discover a table ID such as `tblABC123DEF456` before reading records.",
+      "parameters": {
+        "type": "object",
+        "properties": {
+          "baseId": {
+            "type": "string",
+            "description": "Base ID from `airtable_list_bases`, for example `appABC123DEF456`."
+          }
+        },
+        "required": ["baseId"]
+      },
+      "endpointMapping": {
+        "method": "GET",
+        "path": "/meta/bases/{baseId}/tables"
+      }
+    },
+    {
+      "name": "airtable_list_records",
+      "description": "List records from a table. The response may include an `offset`; pass it to the next call until it is absent. To filter records, use Airtable formula syntax, for example `filterByFormula: {Status}=\"Active\"`.",
+      "parameters": {
+        "type": "object",
+        "properties": {
+          "baseId": {
+            "type": "string",
+            "description": "Base ID, for example `appABC123DEF456`."
+          },
+          "tableIdOrName": {
+            "type": "string",
+            "description": "Table ID or table name, for example `tblABC123DEF456` or `Tasks`. Prefer the table ID returned by `airtable_list_tables`."
+          },
+          "filterByFormula": {
+            "type": "string",
+            "description": "Airtable formula that records must satisfy, for example `{Status}=\"Active\"`. Field names are wrapped in braces."
+          },
+          "view": {
+            "type": "string",
+            "description": "Saved view name or ID to apply its filters and sorting, for example `Active tasks`."
+          },
+          "pageSize": {
+            "type": "integer",
+            "description": "Records per response page, from 1 to 100. For example, `100`."
+          },
+          "maxRecords": {
+            "type": "integer",
+            "description": "Maximum total number of records Airtable should return, for example `250`."
+          },
+          "offset": {
+            "type": "string",
+            "description": "Pagination cursor from the previous response, for example `itrABC123DEF456`. Omit for the first page."
+          }
+        },
+        "required": ["baseId", "tableIdOrName"]
+      },
+      "endpointMapping": {
+        "method": "GET",
+        "path": "/{baseId}/{tableIdOrName}",
+        "queryParams": {
+          "filterByFormula": "$filterByFormula",
+          "view": "$view",
+          "pageSize": "$pageSize",
+          "maxRecords": "$maxRecords",
+          "offset": "$offset"
+        }
+      }
+    },
+    {
+      "name": "airtable_get_record",
+      "description": "Get one record by ID from a table, for example record ID `recABC123DEF456`. Use `airtable_list_records` to find record IDs.",
+      "parameters": {
+        "type": "object",
+        "properties": {
+          "baseId": {
+            "type": "string",
+            "description": "Base ID, for example `appABC123DEF456`."
+          },
+          "tableIdOrName": {
+            "type": "string",
+            "description": "Table ID or table name, for example `tblABC123DEF456` or `Tasks`."
+          },
+          "recordId": {
+            "type": "string",
+            "description": "Record ID, for example `recABC123DEF456`."
+          }
+        },
+        "required": ["baseId", "tableIdOrName", "recordId"]
+      },
+      "endpointMapping": {
+        "method": "GET",
+        "path": "/{baseId}/{tableIdOrName}/{recordId}"
+      }
+    }
+  ]
+}

server.json

@@ -1,7 +1,7 @@
 {
   "$schema": "https://static.modelcontextprotocol.io/schemas/2025-12-11/server.schema.json",
   "name": "io.github.HelpCode-ai/anythingmcp",
-  "description": "189 connectors + any REST/SOAP/GraphQL/SQL as MCP tools for Claude & ChatGPT. Self-hosted, AGPL.",
+  "description": "190 connectors + any REST/SOAP/GraphQL/SQL as MCP tools for Claude & ChatGPT. Self-hosted, AGPL.",
   "repository": {
     "url": "https://github.com/HelpCode-ai/anythingmcp",
     "source": "github"