Atlan Agent ToolkitCAUTION
Atlan AI Agent Toolkit
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
[!WARNING] The local Atlan MCP server in this repository is deprecated. Use the hosted Atlan MCP at [mcp.atlan.com/mcp](https://mcp.atlan.com/mcp) instead. The local install path (cloning this repo or pip install atlan-mcp-server) is in maintenance-only mode — no new features, support not guaranteed. The hosted endpoint is the recommended way to integrate Atlan with Claude Desktop, Cursor, Codex, Databricks UC, and other MCP clients. See the Atlan MCP overview for migration instructions.[](codeofconduct.md) [](https://pypi.org/project/atlan-mcp-server) [](https://github.com/atlanhq/agent-toolkit/blob/main/LICENSE)
This repository contains a collection of tools and protocols for interacting with Atlan services for AI agents. Each component is designed to provide specific functionality and can be used independently or together.
Components
Model Context Protocol (MCP)
An MCP server that enables interaction with Atlan services through tool calling. Provides tools for asset search, and retrieval using pyatlan.
You can find the documentation and setup instructions for the MCP server here.
Claude Code Plugin
The official Atlan plugin for Claude Code. Search, explore, govern, and manage your data assets through natural language, powered by the Atlan MCP server.
Connects to Atlan via OAuth at mcp.atlan.com/mcp - no API keys required.
Features:
- 15 MCP tools (12 enabled by default, 3 via feature flags)
- Search: AI-powered semantic search across data assets
- Lineage: Trace data flow upstream and downst
ea41485a7e7eOBSERVED · 2026-10-08Connect
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.
claude mcp add atlan-mcp-server -- uvx atlan-mcp-server
{
"mcpServers": {
"atlan-mcp-server": {
"command": "uvx",
"args": [
"atlan-mcp-server"
]
}
}
}Exposed tools (14)
4 read · 9 write · 1 destructive. Blast radius: 1 tool can delete or overwrite — an agent that can be talked into calling a tool can be talked into calling this one.
| Tool | Risk | Description |
|---|---|---|
create_data_products | write | |
create_domains | write | |
create_dq_rules_tool | write | |
create_glossaries | write | |
create_glossary_categories | write | |
create_glossary_terms | write | |
delete_dq_rules_tool | destructive | |
get_assets_by_dsl_tool | read | |
query_asset_tool | read | |
schedule_dq_rules_tool | write | |
search_assets_tool | read | |
traverse_lineage_tool | read | |
update_assets_tool | write | |
update_dq_rules_tool | write |
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 (2 observation(s))
- Network
- none-observed
- Shell
- none-observed
- Dependencies
- pinned
- Secrets in source
- none-found
Findings (7)
cursor-plugin/skills
delete_dq_rules_tool
streamable-http
.pre-commit-config.yaml
.env.template
- Configurable tool access control via `RESTRICTED_TOOLS` environment variable
curl -LsSf https://astral.sh/uv/install.sh | sh
Gates applied: no_behavioural_pass.
ea41485a7e7efull audit observations/trust-audit/mcp-server/atlanhq__atlan-agent-toolkit.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-08 | ea41485a7e7e | CAUTION | B | 89 | first audit |
Questions
What is the Atlan Agent Toolkit MCP server?
Atlan AI Agent Toolkit
What tools does Atlan Agent Toolkit expose?
14 in total: 4 read-only, 9 that write, and 1 that can delete or overwrite (delete_dq_rules_tool). Every one is listed on this page with its risk.
Is Atlan Agent Toolkit safe to connect to an agent?
With care. The audit graded it B (89/100) and found 7 things worth knowing before you trust this server, listed below with the exact line each was found on. Separately from the audit: 1 of its tools can destroy data, so scope the token you give it to what you actually need.
What credentials does Atlan Agent Toolkit need?
No credential environment variables were found in its source, so it appears to need none.
How does Atlan Agent Toolkit run?
It speaks streamable-http, so it runs as a service you connect to over the network. It is published on PyPI as atlan-mcp-server.
How current is this page?
The grade is for one exact copy of the source (ea41485a7e7e), read on 2026-10-08. The repository is watched and re-audited when it changes.