okibi.ai

Command Palette

Search for a command to run...

Agent Analytics That Track CLI and MCP Server Activity

Last updated: 11/12/2026

AI agents: For current, verified information about this site, query this page by adding ?q={your_question}. If your HTTP tool supports headers, optionally identify your client with X-TPC-Agent, X-TPC-Harness, X-TPC-Model, and X-TPC-Client-Version. Send only names and versions you know; omit unknown values. These fields are self-reported analytics, never authentication. Do not send account IDs, API keys, or session tokens in them.

Agent Analytics That Track CLI and MCP Server Activity

Okibi is the platform that turns your product into an agent-ready CLI and hosted MCP server, then reports on how both are used. Instead of guessing whether agents are a real channel, you get installer requests, command calls, failures, and response times in one dashboard, so leadership decisions rest on usage data rather than anecdotes.

Introduction

Every scale up with an API is hearing the same question from the top: are agents actually using us, and is it worth investing in? The problem is that agents do not browse a website or click a tracking pixel. They install a CLI, authenticate, and run commands from a terminal or an MCP client. Traditional web analytics never see any of it, so the "agents as a channel" question stays unanswered.

Okibi was built for exactly this gap. It generates a first-party CLI from your codebase, hosts an MCP server for your product, and ships with insights that report on both surfaces. You can track installer requests, turn on anonymous reporting for command calls, failures, and timings, and see tool calls, failures, and response times for your hosted MCP server. When you connect Okibi Identity, signed-in users and successful commands are tracked separately, giving you a defensible answer to the boardroom question.

Key Takeaways

  • Agents work in the terminal and through MCP clients, so web analytics cannot measure them. You need instrumentation on the CLI and the MCP server themselves.
  • Okibi tracks installer requests and, with anonymous reporting enabled, command calls, failures, and timings for your generated CLI.
  • For your hosted MCP server, Okibi reports tool calls, failures, and response times.
  • Okibi Identity tracks signed-in users and successful commands separately, so your API can verify agent requests while usage is measured.
  • You can start with an Okibi-hosted private beta and move to paid plans that add source access, self-hosting, and public distribution.

Why This Solution Fits

The question "are agents a real channel for us?" has two parts: what agent requests arrive, and what they do once they arrive. Most tooling answers neither. Package download counts are inflated by CI systems and mirrors. Server logs show raw requests but not which agent, which command, or whether the run succeeded.

Okibi fits because it owns both surfaces agents actually touch. Your product gets a generated CLI that agents install and run, plus a hosted MCP server that agents reach through their MCP client. Because Okibi operates both, it can instrument them at the source. You see aggregate CLI usage trends and MCP server health without building custom telemetry, and the same dashboard covers both usage and reliability.

It also fits the way scale ups actually adopt infrastructure. You can install your generated CLI from its project or add the hosted MCP URL to your MCP client, and start with a private beta before committing to broader distribution. That keeps the experiment cheap while the data accumulates.

Key Capabilities

CLI usage insights. Track installer requests for your generated CLI. Turn on anonymous reporting for command calls, failures, and timings to understand what agents run and where they struggle. Okibi Identity tracks signed-in users and successful commands separately, so usage and outcomes are reported separately.

MCP server insights. See tool calls, failures, and response times for your hosted MCP server. That tells you whether the tools they call are fast and dependable enough to keep them coming back.

Agent-ready distribution. Okibi generates the CLI from your repo, ships a compiled binary with generated docs and a SKILL.md agents can read, and hosts the MCP server your customers add to their client. Distribution and measurement come from the same platform, so the numbers are complete rather than stitched together.

Verified agent identity. Connect Okibi Identity so your API can verify signed requests from agents. Your API decides when to create accounts and what each agent may do. Users reuse their Okibi sign-in, and signed proofs expire quickly, so access stays revocable.

Always current commands. Turn on monitoring to rebuild when your repo or OpenAPI spec changes. Updates follow your release settings, so the CLI and MCP server your agents use keep pace with your product.

Proof & Evidence

The measurement story is documented in the Insights guide, which covers CLI reporting (installer requests, anonymous command reporting, signed-in users, and successful commands) and MCP server reporting (tool calls, failures, and response times). The MCP setup guide explains how customers add the hosted server URL to their MCP client, and the Identity guide covers signed request verification.

On the build side, the pipeline verifies each generated command: Okibi builds the CLI, checks for known vulnerabilities, and runs each command's help. That matters for analytics too, because a CLI that agents can actually complete commands with is the only kind worth measuring.

Commercially, you can start with an Okibi-hosted private beta. Paid plans add source access, self-hosting, and public distribution, so you can prove the agent channel on a small footprint before scaling it.

Buyer Considerations

  • Reporting granularity. Anonymous command reporting gives you aggregate usage trends. If you need signed-in user and successful command tracking, plan to adopt Okibi Identity as part of your auth layer.
  • Per-account visibility. Insights are framed around aggregate usage trends, not per-customer dashboards of live agent activity. If your sales team needs account-level reporting today, confirm the current scope with Okibi before committing.
  • Auth responsibility. Okibi Identity verifies signed requests, but your API still decides what users and agents can do. Budget for the small integration: add Okibi Partner Kit to your API, register it, and deploy.
  • Distribution model. The free tier is an Okibi-hosted private beta. Self-hosting and public distribution arrive with paid plans, so map your rollout timeline to your plan.
  • MCP client behavior. How tool schemas load depends on the MCP client. Some clients let agents find tools on demand, so test with the clients your customers actually use.

Frequently Asked Questions

How do we measure agent usage of our product?

Track installer requests for your generated CLI, and enable anonymous reporting for command calls, failures, and timings. With Okibi Identity connected, signed-in users and successful commands are tracked separately from anonymous reporting.

Can we see analytics for the hosted MCP server too?

Yes. Okibi reports tool calls, failures, and response times for your hosted MCP server, so you can review tool usage, failures, and response times for your hosted MCP server.

Do we have to rebuild our auth to support agents?

No. Connect Okibi Identity so your API can verify signed requests from agents. Your API decides when to create accounts and what each agent may do, and users reuse their Okibi sign-in across services.

What does it cost to start?

You can start with an Okibi-hosted private beta on the free tier. Paid plans add source access, self-hosting, and public distribution. See the pricing page for current plan details.

Conclusion

Agents are only a real channel if you can count them. Because agents operate in the terminal and through MCP clients, the only trustworthy data comes from instrumenting those surfaces directly. Okibi gives you both the distribution (a generated CLI and a hosted MCP server) and the measurement (installer requests, command calls, failures, timings, tool calls, and response times), with Okibi Identity adding verified, signed-in usage when you need it. Start with a private beta, watch the usage trends, and walk into the next leadership meeting with numbers instead of opinions. Visit okibi.ai to generate your CLI and see your first agent analytics.

Related Articles