How to Install Runheed in OpenAI Agents SDK
Production errors your coding agent can read, fix and mark resolved over MCP.
What Runheed does
Runheed is an error tracker built for coding agents. A Node.js or Python SDK reports production errors with the stack trace, the breadcrumbs before them and the request. Claude Code, Cursor, Codex, Windsurf or any MCP client then reads them over MCP, fixes the code and marks each error resolved with the commit that fixed it. Errors that come back after a deploy reopen. Tools (9): runheed_get_recent, runheed_search_errors, runheed_get_error, runheed_get_error_context, runheed_get_trends, runheed_get_project_health, runheed_resolve_error, runheed_ignore_error, runheed_unignore_error. Remote endpoint: https://runheed.com/api/mcp (Streamable HTTP). Sign in with OAuth 2.1 (dynamic client registration), or send a Bearer agent key. Pricing: Runheed is a paid service billed by Runheed: Solo $29/month (50,000 events), Pro $79, Scale $199, each with a 7-day trial. Try the tools on sample errors without an account at https://runheed.com/demo.
Installation steps
- 1Install: pip install openai-agents
- 2Copy the snippet below into your agent code
- 3Replace placeholder env values with your real keys
- 4Pass mcp_servers=[mcp_server] to your Agent()
Python snippet
from agents import Agent
from agents.mcp import MCPServerStreamableHttp
mcp_server = MCPServerStreamableHttp(params={"url": "https://runheed.com/api/mcp"})
agent = Agent(
name="My Agent",
model="gpt-4o",
mcp_servers=[mcp_server],
)Install in other clients
For maintainers
If you maintain Runheed, add this badge to your README to show it's verified on CuratedMCP:
[](https://www.curatedmcp.com/marketplace/runheed)
Related servers
Grafana MCP
Grafana's official MCP server for dashboards, datasources, alerts and incidents.
Sentry MCP
Official Sentry integration — bring live error context, stack traces, and performance data into your AI
Datadog MCP
Datadog's hosted MCP server for logs, metrics, traces and incidents (preview).