Borrowing it
Nothing to install: this file belongs to comet-ml/opik-mcp. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/comet-ml/opik-mcp/main/AGENTS.mdgit clone --depth 1 https://github.com/comet-ml/opik-mcpWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/comet-ml/opik-mcp/agents-md)<a href="https://agentmods.dev/instructions/comet-ml/opik-mcp/agents-md"><img src="https://agentmods.dev/badge/instructions/comet-ml/opik-mcp/agents-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/comet-ml/opik-mcp/agents-md"><img src="https://agentmods.dev/badge/instructions/comet-ml/opik-mcp/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.01716 | $0.01716 |
| Opus 5.5 | $0.00686 | $0.00686 |
| Sonnet 5.5 | $0.00343 | $0.00343 |
| Haiku 4.5 | $0.00172 | $0.00172 |
Grade A, and why
opik-mcp AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 13d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 129 lines — stays where its author put it; the contents beside it link to each section on GitHub.
opik-mcp
Installing the server to use it? Follow README.md; this file is for changing the code.
IMPORTANT: every token this server puts into a host's context has to pay for itself. Before adding output, a tool, a field or description text, say what it costs (bytes on the surface, tokens in a typical answer) and what the user gets for it. See ADR 0001.
- The advertised surface has a hard byte budget.
- Every
readandlistanswer states its size. - The caller can always narrow: a span, a filter, a window,
fields=[…].
What this is
An MCP server that lets an AI agent (Claude Code, Cursor, claude.ai) work with
a user's Opik data: traces, spans, threads, projects, experiments, datasets,
prompts and Diagnostics issues. The agent can answer "how is my project doing"
or "what broke", and make small changes like scoring or closing a thread. It
also authors the agent skills published as comet-ml/opik-skills.
How it ships: the opik-mcp package on PyPI (stdio, local API key), and a
Docker image served at https://www.comet.com/opik/api/v1/mcp (Streamable
HTTP, the user's OAuth token or API key passed through to the backend).
Stack: Python 3.13, uv, the official MCP python-sdk (FastMCP, pinned
<2), httpx to the Opik REST API, Pydantic.
How a call flows:
read/list:server/tools/→read_list/read_tool.pyorlist_tool.py→ the entity'sEntityHandlerfromread_list/registry.py→ its module inread_list/entities/→client/→ Opik backend.write:writes/write_tool.py→writes/dispatch.py, which runs the operation fromwrites/registry.pythrough its hooks inwrites/operations/→client/→ backend.schema(operation)returns one operation's input on demand.read_skill: servessrc/opik_mcp/skills/fromskills_catalog.py.instructions.pyrenders the text a host receives oninitialize.analytics/sends product events; it is off in tests.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 13d ago Changed · +4 lines · +32 tokens per session b6ca7899b8b8
- 18d ago Changed · +3 lines · +45 tokens per session d05fa5b132f2
- 19d ago Changed · -3 tokens per session 4737c06bccc4
- 20d ago Changed · +4 tokens per session 079a71f5fdfe
- 23d ago Changed · +3 lines · +71 tokens per session 2312fab35692
- 24d ago First seen · 119 lines · 1,567 tokens per session scan A ea2975bcfb3a
opik-mcp AGENTS.md is an instructions file published in the GitHub repository comet-ml/opik-mcp (223 stars, last pushed 2d ago), licensed Apache-2.0. It adds 1,716 tokens to every session, about $0.0069 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-25.
Other instructions, from other repositories
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).
deepseek-harness AGENTS.md
AGENTS.md instructions for deepseek-ai/deepseek-harness, covering agents.md, pre-stable apis and released session data, repository layout, commands and host sandbox failures.