Atlas / MCP servers / rashidazarang / Airtable

AirtableCAUTION

Name: Airtable
Author: rashidazarang

mcp/rashidazarang/airtable-3

MCP server by rashidazarang · airtable · airtable api · api · Audited 2026-10-07

⚑ Report / re-scan Source ↗

Airtable integration for AI-powered applications via Anthropic's Model Context Protocol (MCP)

Verdict

CAUTION

Grade

Trust score

89 /100

Exposed tools

122 96r · 20w · 6d

Transport

streamable-http

License

MIT

Stars

Overview

From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.

[](https://archestra.ai/mcp-catalog/rashidazarang__airtable-mcp) [](https://smithery.ai/server/@rashidazarang/airtable-mcp) [](https://github.com/rashidazarang/airtable-mcp) [](https://www.typescriptlang.org/) [](https://github.com/rashidazarang/airtable-mcp) [](https://github.com/rashidazarang/airtable-mcp) [](https://modelcontextprotocol.io/)

A Model Context Protocol (MCP) server for Airtable with full CRUD operations, schema management, record comments, webhooks, batch operations, governance controls, and AI-powered analytics.

Version 5.1.0 | MCP Protocol 2026-07-28 (stateless core, legacy 2025-era clients still served) | Works with Claude, Codex, Cursor, Windsurf, VS Code, and any MCP client | [](https://www.npmjs.com/package/@rashidazarang/airtable-mcp)

Quick Start (Claude Desktop)

No installation required — just add this to your Claude Desktop config and restart:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json Windows: %APPDATA%\Claude\claude_desktop_config.json

{
"mcpServers": {
"airtable": {
"command": "npx",
"args": ["-y", "@rashidazarang/airtable-mcp"],
"env": {
"AIRTABLE_TOKEN": "YOUR_AIRTABLE_TOKEN",
"AIRTABLE_BASE_ID": "YOUR_BASE_ID"
}
}
}
}

That's it. npx downloads and

Read from source at commit ef100a5a20b7OBSERVED · 2026-10-07

Connect

Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control. Replace the environment placeholders with a token scoped to the least it needs.

claude-code

claude mcp add airtable-mcp --env AIRTABLE_API_KEY=${AIRTABLE_API_KEY} --env AIRTABLE_API_TOKEN=${AIRTABLE_API_TOKEN} --env AIRTABLE_PERSONAL_ACCESS_TOKEN=${AIRTABLE_PERSONAL_ACCESS_TOKEN} --env AIRTABLE_TOKEN=${AIRTABLE_TOKEN} -- npx -y @rashidazarang/[email protected]

claude-desktop

{
  "mcpServers": {
    "airtable-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@rashidazarang/[email protected]"
      ],
      "env": {
        "AIRTABLE_API_KEY": "${AIRTABLE_API_KEY}",
        "AIRTABLE_API_TOKEN": "${AIRTABLE_API_TOKEN}",
        "AIRTABLE_PERSONAL_ACCESS_TOKEN": "${AIRTABLE_PERSONAL_ACCESS_TOKEN}",
        "AIRTABLE_TOKEN": "${AIRTABLE_TOKEN}"
      }
    }
  }
}

Exposed tools (122)

96 read · 20 write · 6 destructive. Blast radius: 6 tools can delete or overwrite — an agent that can be talked into calling a tool can be talked into calling this one.

Tool	Risk	Description
`algorithm`	read	Prediction algorithm to use
`analysis_type`	read	Type of analysis (trends, statistical, patterns, predictive, anomaly_detection, correlation_matrix)
`analyze_data`	read	Advanced AI data analysis with statistical insights, pattern recognition, and predictive modeling
`audit_type`	read	Type of quality check
`auto_fix_suggestions`	read	Include automated fix suggestions (true/false)
`automation_recommendations`	read	Generate intelligent automation suggestions based on workflow patterns and data analysis
`automation_scope`	read	Scope (single_table, multi_table, cross_base, external_integration)
`base_overview`	read	Overview of the base structure and current workflows
`batch_create_records`	write	Create multiple records at once (up to 10)
`batch_delete_records`	destructive	Delete multiple records at once (up to 10)
`batch_update_records`	write	Update multiple records at once (up to 10)
`batch_upsert_records`	write	Update existing records or create new ones based on key fields
`business_context`	read	Business domain context (sales, marketing, operations, finance, customer_success)
`clarifying_questions`	read	Ask clarifying questions if needed
`complexity_tolerance`	read	Acceptable automation complexity (simple, moderate, advanced)
`compliance_requirements`	read	Data compliance needs (gdpr, hipaa, sox, none)
`confidence_level`	read	Statistical confidence level (0.90, 0.95, 0.99)
`confidence_threshold`	read	Minimum confidence for responses
`context_awareness`	read	Use context from previous queries
`context_tables`	read	Tables that might contain relevant data
`create`	write
`create_base`	write	Create a new Airtable base
`create_comment`	write
`create_field`	write	Add a new field to an existing table
`create_record`	write	Create a new record in a table
`create_report`	write	Generate intelligent reports with AI-powered insights, visualizations, and actionable recommendations
`create_table`	write	Create a new table in the base
`create_view`	write	Create a new view for a table
`create_webhook`	write	Create a new webhook for a table
`current_pain_points`	read	Known issues or bottlenecks in current workflow
`data_insights`	read	Discover hidden patterns, correlations, and business insights using advanced AI algorithms
`data_quality_audit`	read	Comprehensive AI-powered data quality assessment with cleansing recommendations
`data_volume`	read	Expected data volume (small, medium, large, enterprise)
`date_field`	read	Field name for date tracking
`delete_comment`	destructive
`delete_field`	destructive	Delete a field from a table (WARNING: This will permanently delete all data in this field)
`delete_record`	destructive	Delete a record from a table
`delete_table`	destructive	Delete a table (WARNING: This will permanently delete all data)
`delete_webhook`	destructive	Delete a webhook
`depth`	read	Analysis depth
`describe`	read
`describe_table`	read	Get detailed information about a specific table including all fields
`existing_base_id`	read	Base ID to analyze existing schema (optional)
`external_factors`	read	External factors to consider
`feature_fields`	read	Fields to use as predictive features
`field_focus`	read	Specific fields to focus the analysis on
`field_mapping`	read	Explicit field mapping (JSON)
`focus_area`	read	Area to focus automation recommendations
`format_preference`	read	Preferred report format
`frequency_patterns`	read	How often tasks are performed
`get_base_schema`	read	Get complete schema information for a base
`get_record`	read	Get a single record by ID
`get_table_views`	read	List all views for a specific table
`get_view_metadata`	read	Get detailed metadata for a specific view
`get_webhook_payloads`	read	Get webhook payload history
`historical_periods`	read	Historical periods for training
`include_confidence`	read	Include confidence scores for answers (true/false)
`include_confidence_intervals`	read	Include confidence intervals
`include_recommendations`	read	Include AI-generated actionable recommendations (true/false)
`insight_depth`	read	Analysis depth (surface, moderate, deep, comprehensive)
`insight_type`	read	Type of insights (correlations, outliers, trends, predictions, segmentation, attribution, churn_analysis)
`integration_capabilities`	read	Available integration tools (zapier, make, custom_api, native_automations)
`integration_needs`	read	External systems to integrate with
`integration_preferences`	read	Preferred integration tools
`list_bases`	read	List all accessible Airtable bases
`list_collaborators`	read	List collaborators and their permissions for the current base
`list_comments`	read
`list_exceptions`	read
`list_field_types`	read	Get a reference of all available Airtable field types and their schemas
`list_governance`	read
`list_records`	read	List records from a specific table
`list_shares`	read	List shared views and their configurations
`list_tables`	read	List all tables in the Airtable base
`list_webhooks`	read	List all webhooks for the base
`model_type`	read	Prediction model (trend_analysis, seasonal_forecast, regression, classification, time_series)
`natural_language_query`	read	Process natural language questions about your data and provide intelligent answers
`optimization_focus`	read	Focus area (automation, data_quality, collaboration, performance, integration, user_experience)
`optimization_goal`	read	Primary optimization goal
`optimize_workflow`	read	AI-powered workflow optimization with automation recommendations and efficiency improvements
`prediction_horizon`	read	Forecast period (next_week, next_month, next_quarter, next_year)
`prediction_periods`	read	Number of periods to predict
`predictive_analytics`	read	Advanced predictive modeling and forecasting using historical Airtable data
`preserve_history`	read	Maintain audit trail of changes (true/false)
`priorities`	read	Design priorities
`quality_dimensions`	read	Quality aspects to check (completeness, accuracy, consistency, validity, uniqueness, timeliness)
`quality_rules`	write	Data quality rules to apply during transformation
`query`	read
`question`	read	Natural language question about your data
`refresh_webhook`	read	Refresh a webhook to extend its expiration
`report_type`	read	Type of report (executive_summary, operational_dashboard, analytical_deep_dive, performance_metrics, predictive_forecast)
`response_format`	read	Desired response format (narrative, data_summary, visualization_suggestion, action_items)
`scale`	read	Expected data scale
`search_records`	read	Search records with filtering and sorting
`severity_threshold`	read	Minimum severity level to report (low, medium, high, critical)
`smart_data_transformation`	read	AI-assisted data transformation, cleaning, and enrichment with intelligent suggestions
`smart_schema_design`	read	AI-assisted database schema optimization and field relationship analysis
`source_table`	read	Source table for transformation
`stakeholder_level`	read	Target audience (executive, manager, analyst, operational)
`status_field`	read	Field name representing workflow status/stage
`table`	read	Table name or ID to analyze
`tables`	read	Comma-separated list of table names to analyze
`target_audience`	read	Primary audience for the report
`target_field`	read	Field to predict or forecast
`target_format`	read	Desired output format or structure
`target_table`	read	Target table name or ID (can be new)
`team_size`	read	Number of users working with this base
`time_dimension`	read	Time field for temporal analysis
`time_period`	read	Time period for analysis (last_7_days, last_30_days, last_quarter, year_to_date, custom)
`transformation_goal`	read	Goal (normalize, standardize, enrich, cleanse, aggregate, pivot)
`transformation_type`	read	Type of transformation
`update`	write
`update_comment`	write
`update_field`	write	Update field properties
`update_record`	write	Update an existing record
`update_table`	write	Update table name or description
`upload_attachment`	write	Upload/attach a file from URL to a record
`upsert`	read
`use_case`	read	Primary use case (crm, project_management, inventory, content_management, hr, finance)
`validation_rules`	write	Validation rules to apply
`whoami`	read
`workflow_description`	read	Description of current manual processes
`workflow_type`	read	Type of workflow to optimize

Trust audit

CAUTIONgrade B · trust 89/100 Install with care. The audit found things worth knowing before you trust its output.

Layer	What it checks	Result
L0	Provenance & inventory	WARN
L1	Static analysis of the code	PASS
L2	Instruction surface (what it tells the agent)	PASS
L3	Class-specific surface	WARN
L4	Behavioural (sandbox)	SKIPPED

What the source does

Filesystem: declared (5 observation(s))
Network: declared (5 observation(s))
Shell: none-observed
Dependencies: not all pinned
Secrets in source: none-found

Findings (15)

MEDIUMInventory / provenance · inv.binary · CWE-1104

docker/Dockerfile.node

Dockerfile.node

Why it matters. a compiled or binary member cannot be reviewed from source

Fix. ship source, or explain the binary in the README

MEDIUMFilesystem / path · mcp.destructive_tools · CWE-22, CWE-59

batch_delete_records, delete_comment, delete_field, delete_record, delete_table, delete_webhook

Why it matters. 6 tool(s) can delete or overwrite

Fix. prefer a read-only mode or scoped tokens; the page states the blast radius

LOWInventory / provenance · inv.hidden_file · CWE-1104

.eslintrc.js

.eslintrc.js

Why it matters. hidden member outside the usual dotfiles

Fix. review its purpose

LOWInformation disclosure · disclose.log_secret · CWE-209, CWE-532

tests/test_mcp_comprehensive.js:71

console.log(`Token: ${TEST_TOKEN.substring(0, 10)}...${TEST_TOKEN.substring(TEST_TOKEN.length - 10)}`);