Opik Diagnose

by comet-ml

Opik Diagnose is an AI Engineering skill for Claude Code, published by comet-ml in opik-mcp.

217 stars35 forkson comet-ml/opik-mcpAdded 2026/09/04Repository updated 2026/09/03

View source on GitHub Download repo ZIP

claude-codegenerative-aillm-observabilitymcpmcp-servermodel-context-protocolopikpython

Install in seconds

Install Opik Diagnose

Copy Opik Diagnose into your Claude Code skills folder. Run the command in your terminal, or review the source on GitHub before installing.

terminal

npx degit https://github.com/comet-ml/opik-mcp/tree/main/src/opik_mcp/skills/opik-diagnose ~/.claude/skills/opik-diagnose

Requires Node.js. Downloads this skill only — not the rest of the repository — into your Claude Code skills folder.

Without Node.js

git clone https://github.com/comet-ml/opik-mcp.git

Clones the whole repository, then copy the skill’s own directory into your skills folder yourself.

In this catalog

Source file: src/opik_mcp/skills/opik-diagnose/SKILL.md in comet-ml/opik-mcp
Installs to: ~/.claude/skills/opik-diagnose
Collection: One of 5 skills cataloged from this repository
Category: AI Engineering — 2970 skills

What Opik Diagnose does

Opik Diagnose surfaces the online Opik traces and diagnostics issues that need attention, ranked by errors, tool-call failures, latency, regressions, and low eval scores. Use it to triage production agent behavior before handing the top trace to Opik Explain.

Opik Diagnose is cataloged under AI Engineering on DirSkills. Opik Diagnose comes from a repository tagged claude-code, generative-ai, llm-observability, mcp and mcp-server.

Documentation

README

Excerpt

Diagnose — Surface the Traces Worth Attention

Definition of done: a ranked shortlist of the online/production traces (and Diagnostics issues) worth attention, each carrying the signal that flagged it and its trace id, scoped to a project and a recent window, and ready to hand to /opik-explain. "Worth attention" means errored, slow, regressed, or low online-eval score — not a dump of every trace, and never offline experiment results. If the project can't be read, stop at the first genuine blocker and return one next step.

Operate: rank by real signal over live data, surface the few things worth a look, hand the top one to /opik-explain — and change no code. This skill is read-only by design.

Inputs

This is the opening of the README. Read the full README on GitHub.

Commands Opik Diagnose provides

Slash commands named in this skill’s SKILL.md, listed in the order they first appear.

/opik-explain
/opik-diagnose
/opik-evaluate
/opik-compare

Frequently asked about Opik Diagnose

What else does comet-ml publish alongside Opik Diagnose?
Opik Diagnose is one of 5 skills that DirSkills catalogs from comet-ml/opik-mcp, the repository it ships in. Its siblings there include Instrument, Opik and Opik Evaluate. Each one is a separate skill with its own page in this directory, installs the same way Opik Diagnose does, and is maintained by comet-ml in that same repository. The rest of the collection is listed on the comet-ml/opik-mcp page.
How does Opik Diagnose compare to other AI Engineering skills?
Opik Diagnose ranks #2620 by stars among the 2970 AI Engineering skills in this catalog. The most-starred ones next to it are Architecture Decision Records, AI-First Engineering and Agentic OS. DirSkills ranks by the star count of the repository each skill ships in, so that order reflects how popular those repositories are rather than any review of Opik Diagnose against them. Open each page to compare what they document and how they install.