# Fixter > Fixter is an AI-native monitoring platform: logs, traces, and metrics, with every > capability also exposed to AI agents as MCP tools. Complete documentation in a single file: https://docs.fixter.dev/llms-full.txt Any page below is also available as raw markdown: append `.md` to its URL. Setting Fixter up? Connect the MCP server at https://mcp.fixter.dev, then follow Get started > Onboarding. On Claude Code, install the plugin and run `/fixter:onboard` rather than wiring OpenTelemetry by hand: it provisions the key and the log bridge for you. Any other agent fetches the key itself over MCP and follows the OpenTelemetry page. ## Get started - [Introduction](https://docs.fixter.dev/getting-started/introduction/): What Fixter is, what it does with your telemetry, and the five steps that make up the platform. - [Onboarding](https://docs.fixter.dev/getting-started/onboarding/): From nothing to logs arriving, then alerts landing in Slack and getting investigated. MCP first, then your agent does the rest. ## Using Fixter - [Telemetry](https://docs.fixter.dev/using-fixter/telemetry/): What Fixter collects and why: logs, traces, and metrics, what each one answers, what you get without writing code, and what happens to it once it arrives. - [Debugging](https://docs.fixter.dev/using-fixter/debugging/): Work out what is happening in production by querying your logs, traces, and metrics from a connected agent, in the app, or in SQL. - [Dashboards](https://docs.fixter.dev/using-fixter/dashboards/): Ask your agent for a dashboard and get back a stable, shareable link to live panels over your logs, traces, and metrics, which it updates in place when you ask for changes. - [Issues](https://docs.fixter.dev/using-fixter/issues/): Exceptions from the telemetry you already send, parsed and grouped into tracked issues, each one investigated automatically and ready to hand to your coding agent. - [Alerts](https://docs.fixter.dev/using-fixter/alerting/): Rules you define in plain language, plus Fixter's own detectors that alert on error bursts and anomalies without configuration. - [Investigations](https://docs.fixter.dev/using-fixter/investigations/): Alerts landing in a Slack channel Fixter watches are investigated automatically: an AI agent queries your telemetry and replies in the thread with root cause, who is affected, and what to do. ## API & MCP testing - [API & MCP testing](https://docs.fixter.dev/api-testing/overview/): Scheduled tests that call your HTTP endpoints and MCP servers from outside, check the answers, and track uptime and latency; what the two kinds of test are for and where each part is documented. - [Creating your first test](https://docs.fixter.dev/api-testing/first-test/): Set up a first API or MCP test, from an agent or in the app, see the save run it once, and read its first scheduled run; the worked example monitors Fixter's own MCP server. - [MCP tests](https://docs.fixter.dev/api-testing/mcp-tests/): How Fixter probes an MCP server the way an agent would, the four cumulative check depths from Uptime to Tool call, the fields each needs, what a credential gate does, and the assertions a failed run reports. - [MCP protocol revisions](https://docs.fixter.dev/api-testing/mcp-protocol-revisions/): The MCP wire generations and dated revisions Fixter can speak, what Any, Newest supported and a pinned revision mean, what a 401, 402 or 403 does at each check depth, how detection works, and how to watch several revisions with pinned tests. - [HTTP tests](https://docs.fixter.dev/api-testing/http-tests/): What an HTTP test sends, the status pattern, latency, header and body assertions it can make, and how it follows redirects. - [Runs, states and uptime](https://docs.fixter.dev/api-testing/runs-and-uptime/): Scheduling limits, the trial run on save, every run outcome and test state, the transition rules, how uptime is computed, which addresses Fixter will not test, what a run detail shows, and where a run's time went, step by step. - [Request variables](https://docs.fixter.dev/api-testing/request-variables/): The variables a test can write in double braces, what each resolves to, where they work, and which fields do not take them. - [Credentials](https://docs.fixter.dev/api-testing/credentials/): The four credential types a test can use, what each one sends, OAuth2 client credentials, why values are write-only, and how to create and delete one. - [Notifications](https://docs.fixter.dev/api-testing/notifications/): How to choose where a test's notifications go, from an agent or in the app, and what each message says when a test goes down, recovers, is rate limited or clears. - [Allow tests through your firewall](https://docs.fixter.dev/api-testing/firewalls/): How to confirm that a CDN, WAF or rate limiter is blocking a test, what to allowlist (User-Agent first, IP address second), the steps for Cloudflare, Vercel and a generic WAF, and what a 429 does to a test. ## Setup - [Overview](https://docs.fixter.dev/setup/overview/): The connections Fixter uses: telemetry in over OpenTelemetry, tools out over MCP, and Slack and GitHub for investigations. - [Claude Code plugin](https://docs.fixter.dev/setup/claude-code-plugin/): Install the Fixter plugin for Claude Code, and the command and skills it adds. - [Send data with OpenTelemetry](https://docs.fixter.dev/setup/opentelemetry/): Get your telemetry flowing into Fixter in three steps, whatever language or platform you run. - [Kubernetes](https://docs.fixter.dev/setup/kubernetes/): Install the Fixter collector Helm chart to ship pod logs, Kubernetes events, and cluster and host metrics, then point your workloads at it for their own traces and logs. - [AWS Firehose](https://docs.fixter.dev/setup/aws-firehose/): Ship CloudWatch Logs and CloudWatch metric streams into Fixter through an Amazon Data Firehose HTTP delivery stream, for AWS-managed services that never run an OpenTelemetry SDK. - [Fixter MCP server](https://docs.fixter.dev/setup/mcp/): Fixter's MCP server gives Claude Code, Cursor, or any MCP client your production logs, traces, metrics, and error issues, plus alert rules, dashboards, and investigations, behind one endpoint you sign in to with your Fixter account. - [Slack](https://docs.fixter.dev/setup/slack/): Connect your Slack workspace, add a channel, invite the Fixter bot, and turn on auto-investigate. One channel carries both alerts and investigations. - [GitHub](https://docs.fixter.dev/setup/github/): Install the Fixter GitHub App so investigations can open pull requests on the affected repository. ## Reference - [Data model](https://docs.fixter.dev/reference/data-model/): What a log, span, and metric look like inside Fixter, where to see each, and how long data is kept. - [MCP tools](https://docs.fixter.dev/reference/mcp/tools/): Every tool the Fixter MCP server advertises, with its description and parameters exactly as the server ships them, grouped the way the app's Tools setting groups them. - [Managing MCP tool list](https://docs.fixter.dev/reference/mcp/managing-tool-list/): Why a 68-tool MCP server eats your fresh session context and makes an agent pick the wrong tool, and how to hide the tools a session does not need from User settings in the app. - [QuerySQL](https://docs.fixter.dev/reference/querysql/): The complete QuerySQL dialect: tables, dynamic attributes, functions, operators, and how it differs from standard SQL.