Connect
Optimize
Secure
Live 52 Actions
Dagster MCP Server
for AI Agents
Connect your AI agent to StackOne's Dagster MCP server and give it 52 MCP tools out of the box. Auth, tool execution, and security all managed.
Dagster MCP Server
Built by 
StackOne
StackOne 
Coverage
52 Agent Actions
Create, read, update, and delete across Dagster — and extend your agent's capabilities with custom actions.
Authentication
Agent Tool Authentication
Per-user OAuth in one call. Your Dagster MCP server gets session-scoped tokens with zero credentials stored on your infra.
Agent Auth →
Security
Agent Protection
Every Dagster tool response scanned for prompt injection in milliseconds — 88.7% accuracy, all running on CPU.
Prompt Injection Defense →
Performance
Max Agent Context. Min Cost.
Free up to 96% of your agent's context window to enhance reasoning and reduce cost, on every Dagster call.
Tools Discovery →What is the Dagster MCP Server?
A Dagster MCP server lets AI agents read and write Dagster data through the Model Context Protocol — Anthropic's open standard for connecting LLMs to external tools. StackOne's Dagster MCP server ships with 52 pre-built actions, fully extensible via the Connector Builder — plus managed authentication, prompt injection defense, observability, and agent execution runtime. Connect it from MCP clients like Claude Desktop, Claude Code, Cursor, Goose, and VS Code, or from agent frameworks like OpenAI Agents SDK, LangChain, and Vercel AI SDK.
All Dagster MCP Tools
Every action from Dagster's API, ready for your agent. Create, read, update, and delete — scoped to exactly what you need.
Runs
- List Runs
List job runs in the current deployment with optional filtering and cursor pagination.
- Get Run
Retrieve a single run by ID, including status, config, tags, and execution stats.
- Delete Run
Permanently delete a run record from storage.
Assets
- List Assets
List asset keys with optional prefix filter and cursor pagination.
- Get Asset
Retrieve one asset by its key.
Asset Nodes
- List Asset Nodes
List asset nodes (assets that have a Dagster definition) with filters.
- Get Asset Node
Retrieve one asset node (definition-backed asset view) by key.
Sensors
- List Sensors
List all sensors in a repository.
- Get Sensor
Retrieve a single sensor by selector.
Schedules
- List Schedules
List all schedules in a repository.
- Get Schedule
Retrieve a single schedule by selector.
Dynamic Partitions
- Add Dynamic Partition
Add a key to a dynamic partition set.
- Delete Dynamic Partitions
Remove keys from a dynamic partition set.
Backfills
- List Backfills
List partition backfills with optional filtering and cursor pagination.
- Get Backfill
Retrieve one backfill by ID.
Other (37)
- List Run IDs
List run IDs only (bulk ID-only listing) with optional filter and cursor pagination.
- Get Run Group
Retrieve a run and all of its re-execution lineage (root + descendants).
- List Runs Feed
List the UI-style runs feed unifying individual runs and backfills.
- List Run Tag Keys
List distinct tag keys observed across all runs.
- List Run Tags
List distinct tag key/value pairs observed across runs.
- Get Run Logs
Stream the event log for a run with cursor pagination.
- Get Job
Retrieve a job (pipeline) by its selector.
- Get Execution Plan
Compute the step plan that would run for a given run config.
- Get Assets Latest Info
Return the latest materialization info for a list of asset keys.
- List Asset Records
List materialized asset records (assets that have at least one materialization on file).
- List Asset Check Executions
List execution history of an asset check.
- List Repositories
List all code repositories loaded in the deployment.
- Get Repository
Retrieve one repository by selector.
- Get Workspace
Retrieve the workspace and its code locations.
- List Code Location Statuses
List the load status of each code location.
- List Top-Level Resources
List the top-level Dagster resources defined in a repository.
- Get Instance
Retrieve Dagster instance/deployment configuration.
- Set Concurrency Limit
Set the global concurrency limit for a given concurrency key.
- Launch Run
Launch a new run of a Dagster job with the supplied run config and tags.
- Launch Multiple Runs
Launch several runs in a single request.
- Re-execute Run
Re-execute an existing run (full re-run or from-failure).
- Terminate Run
Cancel an in-flight run.
- Validate Run Config
Validate a run config against a job before launching.
- Wipe Assets
Delete materialization history for the given asset partition ranges.
- Report Asset Materialization
Report a materialization of an external asset via the External Assets REST API.
- Report Asset Observation
Report an observation of an external asset via the External Assets REST API.
- Report Asset Check
Report the result of an asset quality check via the External Assets REST API.
- Start Sensor
Start (enable) a sensor.
- Stop Sensor
Stop (disable) a running sensor.
- Start Schedule
Start (enable) a schedule.
- Stop Schedule
Stop (disable) a running schedule.
- Launch Backfill
Start a new partition backfill.
- Cancel Backfill
Cancel an in-flight backfill.
- Resume Backfill
Resume a failed backfill.
- Reload Code Location
Reload a single code location.
- Reload Workspace
Reload every code location in the workspace.
- Pause / Resume Auto-Materialize
Pause or resume the global auto-materialize daemon.
Set Up Your Dagster MCP Server in Minutes
One endpoint. Any framework. Your agent is talking to Dagster in under 10 lines of code.
Agent Frameworks
JSON{
"mcpServers": {
"stackone": {
"command": "npx",
"args": [
"-y",
"mcp-remote@latest",
"https://api.stackone.com/mcp?x-account-id=<account_id>",
"--header",
"Authorization: Basic <YOUR_BASE64_TOKEN>"
]
}
}
}Platform Resources
MCP Code Mode: Keeping Tool Responses Out of Agent Context
Anthropic's code_execution processes data already in context. Custom MCP code mode keeps raw tool responses in a sandbox. 14K tokens vs 500.
11 min
Comparing BM25, TF-IDF, and Hybrid Search for MCP Tool Discovery
Benchmarking BM25, TF-IDF, and hybrid search for MCP tool discovery across 916 tools. The 80/20 TF-IDF/BM25 hybrid hits 21% Top-1 accuracy in under 1ms.
10 min
Indirect Prompt Injection Defense for MCP Tools: A Technical Guide
MCP tools that read emails, CRM records, and tickets are indirect prompt injection vectors. Here's how we built a two-tier defense that scans tool results in ~11ms.
12 min
Dagster MCP Server FAQ
Does StackOne have a Dagster MCP server?
Yes. StackOne offers a hosted Dagster MCP server with 52 pre-built actions, and every action is tested and QA'd by StackOne. Connect it to Claude, Cursor, and any other MCP client, or to any agent framework through the AI Action SDK. It ships with managed agent authentication, prompt injection defense, and tool discovery with server-side execution that preserve your agent's context window and keep reasoning performance.
Dagster MCP server vs direct API integration — what's the difference?
A Dagster MCP server and direct API integration serve different use cases. Direct API integration is for software-to-software — backend code calling Dagster. A Dagster MCP server is for AI agents — MCP clients like Claude and Cursor, plus framework agents built with OpenAI, LangChain, or Vercel AI — discovering and calling Dagster at runtime. StackOne provides both.
How does Dagster authentication work for AI agents?
Dagster authentication for AI agents works through a StackOne Connect Session. Create one via the dashboard or the SDK — you get an auth link and ready-to-paste config for Claude Desktop, Cursor, and other MCP clients. Your user authenticates their own Dagster account; StackOne handles token exchange, storage, and refresh. Credentials never reach the LLM, and each user is isolated via
origin_owner_id.Are Dagster MCP tools vulnerable to prompt injection?
Yes — Dagster MCP tools can be vulnerable to indirect prompt injection. Any tool that reads user-written content — documents, messages, tickets, records, or free-text fields — is a potential vector. StackOne Defender scans every tool response before it enters the agent's context — regex patterns in ~1ms, then a MiniLM classifier in ~4ms. 88.7% accuracy, CPU-only.
What is the context bloat of a Dagster agent and how do I avoid it?
Context bloat happens when Dagster tool schemas and API responses eat your Dagster agent's memory, preventing it from reasoning effectively. A single Dagster query can return a massive JSON response, and connecting multiple tools compounds the problem. Tools Discovery and Code Mode reduce context bloat — loading only relevant tools per query and keeping raw responses out of the agent's context.
Can I limit which actions my Dagster agent can access?
Yes — you can limit which actions your Dagster agent can access directly from the StackOne dashboard. Toggle actions on or off, or restrict them to specific accounts, with no code changes to your agent. Session tokens can be scoped to exact actions so if one leaks, exposure stays contained.
Can I create custom agent actions for my Dagster MCP server?
Yes — you can create custom agent actions for your Dagster MCP server using Connector Builder. It's an integration agent your coding assistant (Claude Code, Cursor, or Copilot) can invoke to research Dagster's API, generate production-ready connector YAML, test against the live API, and validate before you ship.
When should I NOT use a Dagster MCP server?
Skip a Dagster MCP server if your integration is purely software-to-software — direct Dagster API integration is simpler when no AI agent is involved. For deterministic, compliance-critical operations (financial transactions, regulatory reporting), direct API gives you predictable behavior without agent-driven decision-making. MCP shines when AI agents need to dynamically discover and call Dagster actions at runtime.
What AI frameworks and AI clients does the StackOne Dagster MCP server support?
The StackOne Dagster MCP server supports both. MCP clients (paste-and-go apps): Claude Desktop, Claude Code, Cursor, VS Code, Goose. Agent frameworks (code SDKs you build with): OpenAI Agents SDK, Anthropic, Vercel AI, Google ADK, CrewAI, Pydantic AI, LangChain, LangGraph, Azure AI Foundry.
Put your AI agents to work
All the tools you need to build and scale AI agent integrations, with best-in-class connectivity, execution, and security.