Opik MCP
Comet (official Opik project)
Comet (official Opik project)
Comet's official MCP integration for Opik lets an AI client inspect LLM traces and projects, record scores and comments, and manage prompts, datasets, and experiments in the connected Opik workspace.
The current official README lists the complete four-tool surface: read, list, write, and schema. These are documentation-derived names, not a live tools/list result. `write` can make changes in the connected Opik workspace; inspect operation schemas and obtain user approval before write operations.
readRead an Opik entity by ID, name, or Opik URI.
listList and filter Opik entities with pagination.
writeCreate or update traces, spans, scores, comments, prompts, datasets, experiments, and selected agent-insights records.
schemaInspect the required shape, scope, and example for supported operations without calling the backend.
This server’s available tools can vary by account, permissions, or configuration. The list shown is not its full runtime catalog.
Identifies the maintained official Opik MCP, distinguishes hosted Cloud OAuth and local stdio paths, documents the hosted endpoint, current uvx package, setup flow, Python requirement, validation command, and the four tools read/list/write/schema.
Current PyPI distribution is opik-mcp and requires Python >=3.13.
Comet organization image linked from the official Comet-owned MCP repository; downloaded, decoded, and reviewed individually and on light/dark contact sheets. Selection tier B: publisher/company logo because no dedicated standalone logo is published in the MCP repository.
Opik MCP appeared in the awesome-mcp.tools observability search results; used for discovery only, with all product and setup facts checked against official Comet sources.
Official repository documents four tools: read/list/write/schema; catalog rows match.
Documentation checked 2026-10-28T00:00:00Z. Connection has not been runtime tested.
Comet (official Opik project)
Comet's official MCP integration for Opik lets an AI client inspect LLM traces and projects, record scores and comments, and manage prompts, datasets, and experiments in the connected Opik workspace.
The current official README lists the complete four-tool surface: read, list, write, and schema. These are documentation-derived names, not a live tools/list result. `write` can make changes in the connected Opik workspace; inspect operation schemas and obtain user approval before write operations.
readRead an Opik entity by ID, name, or Opik URI.
listList and filter Opik entities with pagination.
writeCreate or update traces, spans, scores, comments, prompts, datasets, experiments, and selected agent-insights records.
schemaInspect the required shape, scope, and example for supported operations without calling the backend.
This server’s available tools can vary by account, permissions, or configuration. The list shown is not its full runtime catalog.
Identifies the maintained official Opik MCP, distinguishes hosted Cloud OAuth and local stdio paths, documents the hosted endpoint, current uvx package, setup flow, Python requirement, validation command, and the four tools read/list/write/schema.
Current PyPI distribution is opik-mcp and requires Python >=3.13.
Comet organization image linked from the official Comet-owned MCP repository; downloaded, decoded, and reviewed individually and on light/dark contact sheets. Selection tier B: publisher/company logo because no dedicated standalone logo is published in the MCP repository.
Opik MCP appeared in the awesome-mcp.tools observability search results; used for discovery only, with all product and setup facts checked against official Comet sources.
Official repository documents four tools: read/list/write/schema; catalog rows match.
Documentation checked 2026-10-28T00:00:00Z. Connection has not been runtime tested.