Logs Claude Code sessions to Opik as traces and adds skills for instrumenting and evaluating AI applications.
Also known as comet-ml/opik-claude-code-plugin
- Stars
- 25
- Tools
- —
- Downloads, 30 days
- —
- Latest version
- —
- Last commit
- Oct 22, 2026
- License
- Apache-2.0
What is Opik MCP?
Opik logs Claude Code sessions to Opik as traces and adds skills for instrumenting and evaluating AI applications.
Summary generated from the project's README
About Opik MCP server
Developers using Claude Code can monitor their AI sessions by logging them as traces to the Opik observability platform. This plugin automatically captures conversation turns, tool calls, and thoughts as spans, while also tracking nested subagent calls through parent-child relationships. It is designed to provide visibility into how coding agents behave during complex tasks without slowing down the user experience.
Beyond session logging, the plugin provides specialized skills and agents to help build and maintain observable AI applications. Users can instrument their own Python or TypeScript code by automatically detecting frameworks like LangChain, OpenAI, or LlamaIndex and adding the necessary tracing decorators. The setup requires installing the plugin via the Claude Code marketplace and configuring a connection to Opik using the Opik CLI with an API key and workspace settings.
This tool distinguishes itself by offering a dedicated agent-reviewer to check agent architecture for security and idempotence. It also includes a comprehensive skill pack for tasks such as debugging failed traces, comparing experiment results, and optimizing prompts directly within the AI interface.
What you can do with it
- 1Trace Claude Code sessionsAutomatically log every interaction and tool call within a Claude Code session to the Opik platform for later review.
- 2Instrument application codeAdd observability to existing Python or TypeScript projects by automatically detecting and integrating supported AI frameworks.
Written from the project's README. Every use case is checked against a quote from it.
Configuration
OPIK_CC_PARENT_TRACE_IDOPIK_CC_PROJECTOPIK_CC_ROOT_SPAN_IDOPIK_CC_TRUNCATE_FIELDSOPIK_CC_WORKSPACE
| Name | Description | |
|---|---|---|
| OPIK_CC_PARENT_TRACE_ID | — | |
| OPIK_CC_PROJECT | — | |
| OPIK_CC_ROOT_SPAN_ID | — | |
| OPIK_CC_TRUNCATE_FIELDS | — | |
| OPIK_CC_WORKSPACE | — |
Automated checks
Ran Nov 14, 2026Full report →Not verified yet. A server needs a score of 80, no serious finding, and enough data for us to check. How checks work · What to fix for more points
- Source repository linked and reachablecomet-ml/opik-claude-code-plugin is public and reachablegithub.com/comet-ml/opik-claude-code-plugin ↗Pass
- Package names this repository as its sourceNot checked
- Build provenance (Sigstore attestation)Not checked
- Open source license (OSI)Apache-2.0spdx.org/licenses/Apache-2.0.html ↗Pass
- Commit or release in the last yearLast commit or release 2026-10-22github.com/comet-ml/opik-claude-code-plugin ↗Pass
- OpenSSF ScorecardNot checked
- Package name not close to a popular packageNot checked
- No critical advisory in the package or its dependenciesNot checked
- No hidden instructions to the model0 tool descriptions and the README read; no override phrases, credential paths or hidden charactersPass
- Destructive tools are labelledNot checked
- Remote endpoint answers the MCP handshakeNot checked
README
View on GitHubOpik Claude Code Plugin
Log Claude Code sessions to Opik for LLM observability, plus skills and agents for building observable AI applications.
Opik is the open-source LLM observability and evaluation platform, built by Comet.
Features
- Session Tracing: Automatically log Claude Code sessions as Opik traces
- Span Tracking: Each tool call becomes a span within the trace
- Subagent Support: Nested agent calls are tracked with parent-child relationships
- Skills: Built-in knowledge for LLM observability, tracing, and evaluation
- Agents: Code review agent for agent architecture best practices
How Tracing Works
We trigger tracing for everything done in Claude Code, but don't slow you down.
Each conversation turn becomes an Opik trace. Tool calls, thoughts, and responses become spans. Subagent invocations are nested under their parent Task span.
Installation
From within Claude Code:
/plugin marketplace add comet-ml/opik-claude-code-plugin
Then install the plugin:
/plugin install opik
Local Install (for contributors)
If you've cloned the repo locally, add it as a marketplace and install from there:
/plugin marketplace add /path/to/opik-claude-code-plugin
/plugin install opik
Important: Restart any running Claude Code sessions after installation. Hooks only load when a session starts.
Configuration
Run the Opik CLI to configure your connection:
pip install opik
opik configure
This creates ~/.opik.config with your API URL, key, and workspace.
Optional Environment Variables
export OPIK_CC_PROJECT="my-project" # Project name (default: claude-code)
export OPIK_CC_WORKSPACE="my-workspace" # Workspace override (default: OPIK_WORKSPACE / ~/.opik.config)
export OPIK_CC_TRUNCATE_FIELDS="false" # Don't truncate large fields
All plugin env vars use the OPIK_CC_ prefix to avoid conflicts with standard Opik SDK variables.
OPIK_CC_WORKSPACE and OPIK_CC_PROJECT let you send Claude Code traces to a different
workspace/project than the rest of your Opik setup, without touching the global
OPIK_WORKSPACE / project_name in ~/.opik.config used by the Opik SDK. Both can also be
set in ~/.opik.config via the plugin-scoped cc_workspace and cc_project keys:
[opik]
workspace = my-sdk-workspace # used by the Opik SDK
project_name = my-sdk-project # used by the Opik SDK
cc_workspace = my-cc-workspace # used only by the Claude Code plugin
cc_project_name = my-cc-project # used only by the Claude Code plugin
Note: for backward compatibility, if
cc_project_nameis not set the plugin still falls back to the sharedproject_namekey.
External Trace Linking
Link Claude Code sessions to existing Opik traces (useful for embedding Claude Code in larger workflows):
export OPIK_CC_PARENT_TRACE_ID="your-trace-id" # Attach to existing trace
export OPIK_CC_ROOT_SPAN_ID="your-span-id" # Set parent span for all Claude Code spans
MCP Server Setup
The Opik MCP server provides Claude with tools to interact with your Opik data - query traces, analyze experiments, and access evaluation results directly in conversation.
For Opik Cloud
Add to your ~/.claude.json:
{
"mcpServers": {
"opik": {
"command": "npx",
"args": ["-y", "opik-mcp", "--apiKey", "YOUR_OPIK_API_KEY"]
}
}
}
Replace YOUR_OPIK_API_KEY with your API key from comet.com.
For Self-Hosted Opik
{
"mcpServers": {
"opik": {
"command": "npx",
"args": ["-y", "opik-mcp", "--apiBaseUrl", "http://localhost:5173/api"]
}
}
}
Adjust the apiBaseUrl to match your Opik instance.
Pre-configured Templates
Copy-ready configurations are available in mcp-configs/mcp-servers.json.
Commands
/opik:trace-claude-code - Claude Code Session Tracing
Enable/disable automatic tracing of your Claude Code sessions to Opik.
/opik:trace-claude-code start # Enable tracing for this project
/opik:trace-claude-code start --debug # Enable tracing + debug logging
/opik:trace-claude-code stop # Disable tracing for this project
/opik:trace-claude-code status # Check project + global + effective state
/opik:trace-claude-code start --global # Enable tracing for all projects
/opik:trace-claude-code stop --global # Disable tracing globally
Tracing state is stored in .claude/.opik-tracing-enabled (project) or ~/.claude/.opik-tracing-enabled (global). Resolution order (first match wins):
- Project file containing
off/disabled→ disabled (opt a single project out of a global enable) - Project file present → enabled (
debugcontent also enables debug logging) - Global file present → enabled (same content rules)
- Neither → disabled
So the project setting always takes precedence over the global one. Running stop in a project while global tracing is on writes the off opt-out for that project rather than removing the global enable.
Note: Changes take effect immediately for new conversation turns (no restart needed).
/opik:instrument - Add Observability to Your Code
Automatically detect frameworks in your code and add the correct Opik integration.
/opik:instrument my_agent.py # Add tracing to a specific file
/opik:instrument # Analyze current context and add tracing
Supports automatic detection and integration for:
- Python: OpenAI, Anthropic, LangChain, LlamaIndex, CrewAI, Bedrock, Groq, LiteLLM
- TypeScript: OpenAI, LangChain, Vercel AI SDK
- Custom code: Adds
@opik.trackdecorators orOpikclient usage
The command ensures tracing starts at your entry point (critical for replay capability) and uses the correct span types.
Skills
Opik skills (vendored from comet-ml/opik-mcp)
The plugin ships the Opik skill pack — the same skills published as opik-skills. Claude loads them by task:
| Skill | Use for |
|---|---|
opik |
SDK reference — span types, integrations, threads, the prompt library |
opik-instrument |
"add opik tracing", "instrument my code" — and verify a real trace lands |
opik-diagnose |
"what is broken in production", "which traces need attention" |
opik-explain |
"why did this trace fail", "debug this trace" |
opik-test |
"turn this trace into a test", "make sure this doesn't happen again" |
opik-compare |
"did my fix work", "which cases regressed" |
opik-evaluate |
"evaluate my agent", "build an eval", "write an LLM judge" |
opik-online-eval |
"score production traces", "take this judge live" |
opik-optimize |
"optimize this prompt", "run the prompt optimizer" |
opik-verify |
"is this safe to ship", "go/no-go on this change" |
They are vendored at a pinned opik-mcp commit (see skills/SHARED.md); edit them upstream, not here.
Agents
agent-reviewer
Reviews agent code for:
- Idempotence and retry safety
- Security vulnerabilities
- Architecture patterns
- State management
- Observability hooks
Directory Structure
opik-claude-code-plugin/
├── .claude-plugin/
│ ├── plugin.json # Plugin manifest
│ └── marketplace.json # Marketplace definition
├── hooks/
│ └── hooks.json # Hook configuration
├── scripts/
│ └── opik-logger # Platform selector script
├── bin/
│ └── opik-logger-* # Compiled binaries (darwin/linux, amd64/arm64)
├── src/
│ └── *.go # Go source code
├── skills/ # Opik skill pack, vendored from comet-ml/opik-mcp (see skills/SHARED.md)
│ ├── opik/
│ ├── opik-instrument/ … opik-optimize/
├── agents/
│ └── agent-reviewer.md # Agent review agent
├── commands/
│ ├── trace-claude-code.md # /opik:trace-claude-code command
│ └── instrument.md # /opik:instrument command
└── mcp-configs/
└── mcp-servers.json # MCP server configurations
Building from Source
make build # Build for all platforms
make build-local # Build for current platform only
Opik for Claude Code
This repo is part of a set of tools for observing Claude Code and other coding agents with Opik:
- opik-claude-code-plugin: log Claude Code sessions as Opik traces, with skills and agents included (this repo)
- ccsync: export Claude Code conversation history to Opik
- cost-intelligence-proxy: meter Claude Code token spend and cost per call
- opik-skills: agent skills for instrumenting your code with Opik
Opik MCP questions
How do I install the Opik plugin for Claude Code?
Add the marketplace and install the plugin from within Claude Code, then restart any running sessions.
How does Opik trace Claude Code sessions?
Each conversation turn becomes an Opik trace, and tool calls, thoughts, and responses become spans, with subagents nested under their parent Task span.
How do I configure the Opik connection?
Install the Opik CLI and run opik configure, which writes your API URL, key, and workspace to ~/.opik.config.
Can I send Claude Code traces to a different project or workspace?
Yes, set OPIK_CC_PROJECT and OPIK_CC_WORKSPACE, or the cc_project_name and cc_workspace keys in ~/.opik.config.
How do I add Opik tracing to my own agent code?
Run /opik:instrument, which detects frameworks like OpenAI, LangChain, and CrewAI and adds the correct Opik integration.