ChatGPT
OpenAIThe default general-purpose assistant, now with agents and deep research.
Pulls LangSmith traces and runs: inputs, outputs, latency, and errors for a project. It turns 'the agent is flaky' into a specific failing step, which is the only way an evaluation loop improves. Read-only…
Reading traces and evaluations to find where an agent fails
Running experiments — that is the platform's job
Pulls LangSmith traces and runs: inputs, outputs, latency, and errors for a project. It turns ‘the agent is flaky’ into a specific failing step, which is the only way an evaluation loop improves. Read-only access is enough for that.
uvx langsmith-mcpPaste this into your MCP client config, then restart the client.
{
"mcpServers": {
"avbbc": {
"command": "npx",
"args": ["-y", "uvx langsmith-mcp"]
}
}
}Connects an agent to a Grafana instance so it can list dashboards, run datasource queries, and inspect alert rules. The value is in incident work: an agent that can …
Pulls issues, stack traces, and event context into the agent so it can triage production errors with the actual payload in front of it. Remote-hosted, which means no…
Queries the Datadog API for metrics, monitors, dashboards, and incident data. On an incident this turns 'something is slow' into an actual series with a timestamp. C…
The default general-purpose assistant, now with agents and deep research.
AI-first workflow builder: nodes that call models, not just apps.
Autonomous coding agent that works in its own sandbox.
Score a search result set for relevance before an agent trusts it.
Turn one question into three queries that cover different phrasings.
Scores retrieval quality on a fixed question set.