> 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/overview.md).

# Overview

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

An AI agent that works on your app needs to see what the app did, not only an error message. Currents gives the agent that data, and gives you links to check the agent's work.

If a CI test covers the change, start with [Agentic testing](/ai/agentic-testing.md): the agent works from the runs Currents already recorded, with their errors, steps, console and network logs, screenshots, videos, traces and history. If no test covers it, start with [Agent evidence](/ai/agent-evidence.md): the agent records what it did as a [session](/ai/reference/sessions.md), and gives you one link to review it.

## Try it in two minutes

1. Connect your agent to the Currents MCP server. See [Connect your agent](/ai/connect.md).
2. Paste a prompt with a run from your project: "Tests are failing in CI. Get the details for run `<runId>` and fix the failures." The run ID is in the run's URL in the dashboard, after `/run/`.
3. The agent reads the failure from Currents, fixes the cause, and runs the test again. See it in [Fix failing CI tests](/ai/agentic-testing/fix-failing-tests.md#example-an-agent-fixes-a-failed-run).

## What do you need?

| You want to                                                                  | Do you have a test for it? | Read                                                                    |
| ---------------------------------------------------------------------------- | -------------------------- | ----------------------------------------------------------------------- |
| Fix a failing test, with its full failure context and history                | Yes                        | [Fix failing CI tests](/ai/agentic-testing/fix-failing-tests.md)        |
| Prove a change works, from the CI runs before and after                      | Yes                        | [Prove a change with CI tests](/ai/agentic-testing/prove-a-change.md)   |
| Share the recording of a CI run with a colleague or an agent                 | Yes                        | [Share a CI run](/ai/agentic-testing/share-a-run.md)                    |
| Prove a change works, with a recording of the browser before and after       | No                         | [Record a bug and its fix](/ai/agent-evidence/bug-and-fix.md)           |
| Keep the logs and files of a task that is not testing                        | No                         | [Record any agent task](/ai/agent-evidence/any-agent-task.md)           |
| Let an agent read a Playwright trace in small steps                          | Either                     | [Explore a trace](/ai/agentic-artifacts/explore-a-trace.md)             |
| Post what Currents recorded on a pull request, in Slack or on a Linear issue | Either                     | [Post to GitHub, Slack, Linear](/ai/agentic-artifacts/post-evidence.md) |
| Give the agent the steps for one task, so the prompt stays short             | Either                     | [Agent skills](/ai/agent-skills.md)                                     |
| Find the flakiest and slowest tests, and when a test started failing         | Either                     | [Connect your agent](/ai/connect.md#try-it), then ask                   |
| Look up links, sessions, trace link routes or MCP tools                      | Either                     | [Reference](/ai/reference.md)                                           |

Agents connect in Claude Code, Cursor, Claude, Codex, ChatGPT, your editor, or any other MCP client. See [Connect your agent](/ai/connect.md).


---

# 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/overview.md?ask=<question>&goal=<user_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.