# Subtext > Session replay, built for agents. Subtext captures every production session cheaply, then gives the coding agent you already run the tools to open one, inspect the moment it broke, and find the root cause. Evidence an agent acts on, not a video a human watches. Subtext is a standalone, self-serve product from Fullstory. You install a capture snippet, connect your agent over MCP or the CLI, and your agent reviews real production sessions. Free to start: no credit card, no seats, no feature gates. Subtext is not a coding agent. You keep the harness you already run (Claude Code, Cursor, Codex, Devin, or your own) and Subtext supplies the session context it has been missing. The category is Agentic Session Review. Install with `npx @subtextdev/subtext-wizard`. Sign up at https://app.fullstory.com/subtext/signup. The free plan is 10,000 captured sessions and 100 agent credits a month, with no credit card. Agents connect to the MCP server at https://api.fullstory.com/mcp/subtext, which requires authentication. The install wizard configures it, and https://subtext.fullstory.com/docs/install/overview covers doing it by hand. For the whole site as one file, read https://subtext.fullstory.com/llms-full.txt. The documentation is a separate set with its own indexes: https://subtext.fullstory.com/docs/llms.txt and https://subtext.fullstory.com/docs/llms-full.txt. Every blog post below is also available as clean markdown at its URL with `.md` appended. ## Product - [Home](https://subtext.fullstory.com): what Agentic Session Review is, the capture-link-review workflow, and which agents it works with. - [Pricing](https://subtext.fullstory.com/pricing): Free $0, Starter $50/mo, Team $250/mo. Two meters: captured sessions and agent credits. A full agent session review costs 10 credits, a quick summary costs 1. - [Install](https://subtext.fullstory.com/install): add capture to your app, then connect the agent that will read the sessions. - [Setup wizard](https://subtext.fullstory.com/wizard): what `npx @subtextdev/subtext-wizard` does, step by step. - [Playground](https://subtext.fullstory.com/playground): watch an agent review a real session in the browser, with nothing installed. - [Blog](https://subtext.fullstory.com/blog): product thinking, research, and company posts. - [Support](https://subtext.fullstory.com/support): chat with support, or email subtext@fullstory.com. ## Documentation - [Introduction](https://subtext.fullstory.com/docs/introduction): What Subtext is and what the agent gets from a session. - [Quickstart](https://subtext.fullstory.com/docs/quickstart): From zero to an agent reviewing a session. - [Install overview](https://subtext.fullstory.com/docs/install/overview): The two halves of setup: the capture snippet in your app, the MCP server in your agent. - [Capture snippet](https://subtext.fullstory.com/docs/install/snippet): What the snippet loads, what it captures, and what it masks by default. - [Manual install](https://subtext.fullstory.com/docs/install/manual): Snippet, plugin, and MCP server wired by hand, for custom frameworks and monorepos. Per-framework guides cover Next.js App Router, React, Vite, and Remix. - [Session review](https://subtext.fullstory.com/docs/session-review/overview): How an agent opens a session, moves through it, and gathers evidence. - [Session review tools reference](https://subtext.fullstory.com/docs/session-review/tools-reference): Every MCP tool an agent can call against a session, with arguments and return shapes. - [Session review skills](https://subtext.fullstory.com/docs/session-review/skills): Packaged agent workflows for reviewing a session and reporting back. - [Privacy](https://subtext.fullstory.com/docs/privacy/overview): Default masking, code-first redaction, and agent-proposed rules a human approves. - [Integrations](https://subtext.fullstory.com/docs/integrations/overview): Attach the session URL to the tools you already run, so any signal links to the real session. 21 guides, from Sentry and Datadog to PostHog, Segment, and LangSmith. - [Recipe: Linear triage](https://subtext.fullstory.com/docs/recipes/linear): Run an agent session review on every issue that enters triage. - [Recipe: LangSmith trace review](https://subtext.fullstory.com/docs/recipes/langsmith-trace-review): Go from a flagged agent run to the session the user actually had. - [Client API](https://subtext.fullstory.com/docs/api/get-session): Get a link to the current session, identify users, and set privacy classes from your own code. - [Verify (closed beta)](https://subtext.fullstory.com/docs/verify/overview): Hosted-browser tooling for agent-driven UI verification. Beta, separate from session review. - [Documentation index](https://subtext.fullstory.com/docs/llms.txt): every documentation page with a one-line summary, generated from the docs site. - [Documentation, full text](https://subtext.fullstory.com/docs/llms-full.txt): the entire documentation set as one file. ## Blog - [Attention Is a Zoom](https://subtext.fullstory.com/blog/attention-is-a-zoom): Every question you ask a session collapses to the same handful of moves: map, then zoom. - [A map, not a movie](https://subtext.fullstory.com/blog/a-map-not-a-movie): A good engineer never watches a session replay end to end. They orient, zoom, and inspect one layer at a time, the way you use the map in your pocket. Agentic Session Review gives your agent that same map. - [Introducing Subtext: Session Replay → Agentic Session Review](https://subtext.fullstory.com/blog/introducing-subtext): Today we're launching Subtext and opening the era of Agentic Session Review. For a decade Fullstory captured behavioral truth so humans could watch replays. It turns out that data is even more valuable to the agents you already use. - [Build Together Again](https://subtext.fullstory.com/blog/build-together-again): AI gave everyone the power to build software. It forgot to give them a place to build it together. - [Your agent can't see what it built](https://subtext.fullstory.com/blog/agents-cant-see): Coding agents write code, spin up previews, and ship PRs. But they still can't perceive their own work, and the humans reviewing it have no good way to talk back. Agentic Session Review fixes both: sessions your agent can read. ## Optional - [Brand](https://subtext.fullstory.com/brand): logos, colors, and usage rules. - [Terms of Service](https://subtext.fullstory.com/tos) - [Privacy Policy](https://www.fullstory.com/legal/privacy-policy/) - [Fullstory](https://www.fullstory.com): the company behind Subtext.