> For the complete documentation index, see [llms.txt](https://docs.currents.dev/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.currents.dev/ai.md). # AI - [Overview](https://docs.currents.dev/ai/overview.md): Give AI agents your Playwright results from CI to fix tests and prove changes, and have agents record screenshots, video and traces of their own work - [Agentic testing](https://docs.currents.dev/ai/agentic-testing.md): AI agents working with your CI test runs in Currents: fix failing tests, prove a change with the tests that cover it, and share a run - [Fix failing CI tests](https://docs.currents.dev/ai/agentic-testing/fix-failing-tests.md): Hand a failing test to an AI agent with the full failure context, from the dashboard, Slack, the GitHub run comment, your editor or the MCP server - [Prove a change with CI tests](https://docs.currents.dev/ai/agentic-testing/prove-a-change.md): Prove an AI agent's change works with the CI tests that cover it: before and after screenshots, videos and traces from the runs Currents already stores - [Share a CI run](https://docs.currents.dev/ai/agentic-testing/share-a-run.md): Share a recording of a CI run with a colleague or an agent: one link to every test, with its screenshots, videos, traces and other files - [Agent evidence](https://docs.currents.dev/ai/agent-evidence.md): Record screenshots, video, a Playwright trace and logs of what your AI agent did, and share them as one link: proof for a pull request, a bug report or a task that is not a test - [Record a bug and its fix](https://docs.currents.dev/ai/agent-evidence/bug-and-fix.md): Have an AI agent record a bug and its fix in the browser, with screenshots, page snapshots and a Playwright trace, and post the before and after on the pull request as proof the change works - [With Playwright MCP](https://docs.currents.dev/ai/agent-evidence/bug-and-fix/playwright-mcp.md): Record what an agent did with Playwright MCP, with a trace, screenshots and page snapshots, and share it as a Currents session - [With Playwright CLI](https://docs.currents.dev/ai/agent-evidence/bug-and-fix/playwright-cli.md): Record what an agent did with Playwright CLI, with a trace, video, screenshots and page snapshots, and share it as a Currents session - [Record any agent task](https://docs.currents.dev/ai/agent-evidence/any-agent-task.md): Record the logs, output files and screenshots of any task an AI agent did, not only testing, and share them as one link that outlives the machine it ran on - [Record local test results](https://docs.currents.dev/ai/agent-evidence/local-test-results.md): Share the traces, screenshots and videos of a Playwright Test or Vitest run that is not reported to Currents, as one session - [Record without the Currents CLI](https://docs.currents.dev/ai/agent-evidence/mcp-tools.md): Record a session with the Currents MCP tools when the agent cannot run the currents command - [Agentic artifacts](https://docs.currents.dev/ai/agentic-artifacts.md): Let agents read Playwright traces in small steps, and post screenshots, traces and recordings from Currents to GitHub, Slack and Linear - [Explore a trace](https://docs.currents.dev/ai/agentic-artifacts/explore-a-trace.md): Give an agent a trace API link, and let it read a Playwright trace in small steps instead of loading the whole file. Get images of the trace for people: filmstrip, animation, frames and the page at th - [Post to GitHub, Slack, Linear](https://docs.currents.dev/ai/agentic-artifacts/post-evidence.md): Have an AI agent post what Currents recorded, such as a session, a shared run, a Fix link, or one screenshot, video or trace image from them, on a GitHub pull request, in Slack or on a Linear issue - [Agent skills](https://docs.currents.dev/ai/agent-skills.md): The agent skills Currents provides, what each is for, and how to make an agent use one - [Currents MCP server skills](https://docs.currents.dev/ai/agent-skills/agent-skill-currents-mcp.md): The skills the Currents MCP server gives a connected agent: fix-failing-tests, collect-evidence and browser-evidence - [Currents CLI skill](https://docs.currents.dev/ai/agent-skills/agent-skill-currents-cli.md): The currents-cli skill teaches an agent with a shell to record sessions and use the currents command - [Playwright best practices](https://docs.currents.dev/ai/agent-skills/agent-skill-playwright-best-practices.md): An agent skill that teaches AI agents to write, debug and maintain Playwright tests the way an experienced Playwright engineer would - [Connect your agent](https://docs.currents.dev/ai/connect.md): Connect Claude, Cursor, Codex, ChatGPT, your editor or any other MCP client to Currents, and pick the setup that fits where you work - [Claude Code](https://docs.currents.dev/ai/connect/claude-code.md): Connect Claude Code to Currents with one command and a browser sign-in, for yourself or for everyone working in a repository - [Cursor](https://docs.currents.dev/ai/connect/cursor.md): Connect Cursor to the Currents MCP server with an install link and a browser sign-in, for yourself or for everyone working in a repository - [Claude Desktop and Claude.ai](https://docs.currents.dev/ai/connect/claude.md): Connect Currents to Claude Desktop and Claude.ai from the Claude connectors directory, for yourself or for your whole Claude organization - [Codex](https://docs.currents.dev/ai/connect/codex.md): Connect Codex (the CLI, the IDE extension and the ChatGPT desktop app) to Currents with one command and a browser sign-in - [ChatGPT](https://docs.currents.dev/ai/connect/chatgpt.md): Add Currents to ChatGPT as a developer-mode app connected to the Currents MCP server - [IDE Extension](https://docs.currents.dev/ai/connect/ide-extension.md): See CI runs and failures in VS Code or Cursor, and send a failure to your AI agent with its full context - [Any MCP client](https://docs.currents.dev/ai/connect/remote.md): Connect any MCP client to the Currents MCP server with a sign-in or an API key, what a connection can reach, and the endpoint reference - [Local MCP Server](https://docs.currents.dev/ai/connect/local.md): Run the Currents MCP server on your own machine with the @currents/mcp package and an API key - [Reference](https://docs.currents.dev/ai/reference.md): Reference for Currents share links, sessions, the trace API, and the MCP server tools and permissions - [Share links](https://docs.currents.dev/ai/reference/share-links.md): What a share link opens when you share a failing test, a run, a browser session, a file or a trace, who can open it, how long it lasts, and how to turn share links off - [Sessions](https://docs.currents.dev/ai/reference/sessions.md): What a Currents session holds, which keys can create and read one, how to list and delete sessions with the API, and their limits - [Trace API](https://docs.currents.dev/ai/reference/trace-api.md): Every route of the Currents trace API, with its options and real output: the trace summary, the player, screen frames, filmstrips and animations, the page at any time in five formats, network requests - [MCP tools and permissions](https://docs.currents.dev/ai/reference/mcp-server.md): Every tool of the Currents MCP server, what it does and the permission it needs, and how the hosted and local servers differ --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation by asking a question. Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters: ``` GET https://docs.currents.dev/ai.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.