Skip to content

Conversation

@miguelg719
Copy link
Collaborator

@miguelg719 miguelg719 commented Dec 5, 2025

why

Anthropic agents in CUA mode are unable to issue key presses (not to be confused with type actions)

what changed

The format for the anthropic tool computer_20250124 replies with:

{
  "action":"key",
  "text":"BackSpace"
}

wasn't properly mapped to our internal action abstraction: keypress, which accepts parameter keys. It was issued directly from the anthropic format. Updated AnthropicCUAClient.ts to account for this and map appropriately

test plan

  • Tested on sample eval

Summary by cubic

Fixes key action mapping in Anthropic CUA so agents can send key presses (e.g., Backspace) correctly instead of failing on the "key" action.

  • Bug Fixes
    • Map Anthropic "key" to internal "keypress" and pass keys from input.text.
    • Remove the old "key" path and Playwright key mapping to avoid mismatches.

Written for commit b9716b9. Summary will update automatically on new commits.

@changeset-bot
Copy link

changeset-bot bot commented Dec 5, 2025

🦋 Changeset detected

Latest commit: b9716b9

The changes in this PR will be included in the next version bump.

This PR includes changesets to release 2 packages
Name Type
@browserbasehq/stagehand Patch
@browserbasehq/stagehand-evals Patch

Not sure what this means? Click here to learn what changesets are.

Click here if you're a maintainer who wants to add another changeset to this PR

@miguelg719 miguelg719 marked this pull request as ready for review December 5, 2025 21:17
Copy link
Contributor

@cubic-dev-ai cubic-dev-ai bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

No issues found across 2 files

@greptile-apps
Copy link
Contributor

greptile-apps bot commented Dec 5, 2025

Greptile Overview

Greptile Summary

Fixed Anthropic CUA agents to properly handle key press actions by mapping the key action type to the internal keypress action format.

  • Removed old key action handler that used mapKeyToPlaywright utility and standalone processing
  • Added key action support to the unified keypress handler, mapping input.text to the keys parameter
  • Removed unused mapKeyToPlaywright import from AnthropicCUAClient

This change enables Anthropic agents in CUA mode to issue key presses (like Backspace, Enter, etc.) which were previously failing.

Confidence Score: 4/5

  • This PR fixes a bug and is mostly safe to merge with one type casting issue to address
  • The core fix correctly maps Anthropic's "key" action to the internal "keypress" action. However, there's a type casting issue on line 777 where input.text (a string) is cast to string[], which could cause the handler to iterate over characters instead of treating it as a single key.
  • Pay close attention to packages/core/lib/v3/agent/AnthropicCUAClient.ts line 777 for the type casting fix

Important Files Changed

File Analysis

Filename Score Overview
.changeset/shy-sides-crash.md 5/5 Standard changeset file for tracking the patch release
packages/core/lib/v3/agent/AnthropicCUAClient.ts 4/5 Fixed key action mapping to properly convert Anthropic's "key" action to internal "keypress" action, but has a potential type casting issue

Sequence Diagram

sequenceDiagram
    participant Agent as Anthropic Agent
    participant Client as AnthropicCUAClient
    participant Handler as v3CuaAgentHandler
    participant Browser as Playwright Page

    Agent->>Client: Tool use: {action: "key", text: "BackSpace"}
    Client->>Client: convertToolUseToAction()
    Note over Client: Maps "key" action to "keypress"
    Note over Client: Extracts text parameter
    Client->>Client: Return {type: "keypress", keys: text}
    Client->>Handler: Execute keypress action
    Handler->>Handler: Normalize keys to array
    Handler->>Handler: Map key to Playwright format
    Handler->>Browser: page.keyPress(mapped)
    Browser-->>Handler: Key pressed
    Handler-->>Client: Action complete
    Client-->>Agent: Screenshot with result
Loading

Copy link
Contributor

@greptile-apps greptile-apps bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

1 file reviewed, 1 comment

Edit Code Review Agent Settings | Greptile

Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com>
@miguelg719 miguelg719 merged commit 6aa9d45 into main Dec 8, 2025
17 checks passed
miguelg719 pushed a commit that referenced this pull request Dec 13, 2025
This PR was opened by the [Changesets
release](https://github.com/changesets/action) GitHub action. When
you're ready to do a release, you can merge this and the packages will
be published to npm automatically. If you're not ready to do a release
yet, that's fine, whenever you add more changesets to main, this PR will
be updated.


# Releases
## @browserbasehq/[email protected]

### Patch Changes

- [#1388](#1388)
[`605ed6b`](605ed6b)
Thanks [@miguelg719](https://github.com/miguelg719)! - Fix multiple
click event dispatches on CDP and Anthropic CUA handling (double clicks)

- [#1400](#1400)
[`34e7e5b`](34e7e5b)
Thanks [@seanmcguire12](https://github.com/seanmcguire12)! - don't write
base64 encoded screenshots to disk when caching agent actions

- [#1345](#1345)
[`943d2d7`](943d2d7)
Thanks [@tkattkat](https://github.com/tkattkat)! - Add support for
aborting / stopping an agent run & continuing an agent run using
messages from prior runs

- [#1334](#1334)
[`0e95cd2`](0e95cd2)
Thanks [@tkattkat](https://github.com/tkattkat)! - Add support for
google vertex provider

- [#1410](#1410)
[`d4237e4`](d4237e4)
Thanks [@seanmcguire12](https://github.com/seanmcguire12)! - fix:
include extract in stagehand.history()

- [#1315](#1315)
[`86975e7`](86975e7)
Thanks [@tkattkat](https://github.com/tkattkat)! - Add streaming support
to agent through stream:true in the agent config

- [#1304](#1304)
[`d5e119b`](d5e119b)
Thanks [@miguelg719](https://github.com/miguelg719)! - Add support for
Microsoft's Fara-7B

- [#1346](#1346)
[`4e051b2`](4e051b2)
Thanks [@seanmcguire12](https://github.com/seanmcguire12)! - fix: don't
attach to targets twice

- [#1327](#1327)
[`6b5a3c9`](6b5a3c9)
Thanks [@miguelg719](https://github.com/miguelg719)! - Informed error
parsing from api

- [#1335](#1335)
[`bb85ad9`](bb85ad9)
Thanks [@seanmcguire12](https://github.com/seanmcguire12)! - add support
for page.addInitScript()

- [#1331](#1331)
[`88d28cc`](88d28cc)
Thanks [@seanmcguire12](https://github.com/seanmcguire12)! - fix:
page.evaluate() now works with scripts injected via
context.addInitScript()

- [#1316](#1316)
[`45bcef0`](45bcef0)
Thanks [@tkattkat](https://github.com/tkattkat)! - Add support for
callbacks in stagehand agent

- [#1374](#1374)
[`6aa9d45`](6aa9d45)
Thanks [@miguelg719](https://github.com/miguelg719)! - Fix key action
mapping in Anthropic CUA

- [#1330](#1330)
[`d382084`](d382084)
Thanks [@seanmcguire12](https://github.com/seanmcguire12)! - fix: make
act, extract, and observe respect user defined timeout param

- [#1336](#1336)
[`1df08cc`](1df08cc)
Thanks [@tkattkat](https://github.com/tkattkat)! - Patch agent on api

- [#1358](#1358)
[`2b56600`](2b56600)
Thanks [@tkattkat](https://github.com/tkattkat)! - Add support for 4.5
opus in cua agent

## @browserbasehq/[email protected]

### Patch Changes

- [#1364](#1364)
[`ca0630e`](ca0630e)
Thanks [@tkattkat](https://github.com/tkattkat)! - Update model handling
in agent evals cli

- Updated dependencies
\[[`605ed6b`](605ed6b),
[`34e7e5b`](34e7e5b),
[`943d2d7`](943d2d7),
[`0e95cd2`](0e95cd2),
[`d4237e4`](d4237e4),
[`86975e7`](86975e7),
[`d5e119b`](d5e119b),
[`4e051b2`](4e051b2),
[`6b5a3c9`](6b5a3c9),
[`bb85ad9`](bb85ad9),
[`88d28cc`](88d28cc),
[`45bcef0`](45bcef0),
[`6aa9d45`](6aa9d45),
[`d382084`](d382084),
[`1df08cc`](1df08cc),
[`2b56600`](2b56600)]:
    -   @browserbasehq/[email protected]

## @browserbasehq/[email protected]

### Patch Changes

- Updated dependencies
\[[`605ed6b`](605ed6b),
[`34e7e5b`](34e7e5b),
[`943d2d7`](943d2d7),
[`0e95cd2`](0e95cd2),
[`d4237e4`](d4237e4),
[`86975e7`](86975e7),
[`d5e119b`](d5e119b),
[`4e051b2`](4e051b2),
[`6b5a3c9`](6b5a3c9),
[`bb85ad9`](bb85ad9),
[`88d28cc`](88d28cc),
[`45bcef0`](45bcef0),
[`6aa9d45`](6aa9d45),
[`d382084`](d382084),
[`1df08cc`](1df08cc),
[`2b56600`](2b56600)]:
    -   @browserbasehq/[email protected]

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
michaelfp930-WB added a commit to michaelfp930-WB/stagehand that referenced this pull request Jan 12, 2026
This PR was opened by the [Changesets
release](https://github.com/changesets/action) GitHub action. When
you're ready to do a release, you can merge this and the packages will
be published to npm automatically. If you're not ready to do a release
yet, that's fine, whenever you add more changesets to main, this PR will
be updated.


# Releases
## @browserbasehq/[email protected]

### Patch Changes

- [#1388](browserbase/stagehand#1388)
[`605ed6b`](browserbase/stagehand@605ed6b)
Thanks [@miguelg719](https://github.com/miguelg719)! - Fix multiple
click event dispatches on CDP and Anthropic CUA handling (double clicks)

- [#1400](browserbase/stagehand#1400)
[`34e7e5b`](browserbase/stagehand@34e7e5b)
Thanks [@seanmcguire12](https://github.com/seanmcguire12)! - don't write
base64 encoded screenshots to disk when caching agent actions

- [#1345](browserbase/stagehand#1345)
[`943d2d7`](browserbase/stagehand@943d2d7)
Thanks [@tkattkat](https://github.com/tkattkat)! - Add support for
aborting / stopping an agent run & continuing an agent run using
messages from prior runs

- [#1334](browserbase/stagehand#1334)
[`0e95cd2`](browserbase/stagehand@0e95cd2)
Thanks [@tkattkat](https://github.com/tkattkat)! - Add support for
google vertex provider

- [#1410](browserbase/stagehand#1410)
[`d4237e4`](browserbase/stagehand@d4237e4)
Thanks [@seanmcguire12](https://github.com/seanmcguire12)! - fix:
include extract in stagehand.history()

- [#1315](browserbase/stagehand#1315)
[`86975e7`](browserbase/stagehand@86975e7)
Thanks [@tkattkat](https://github.com/tkattkat)! - Add streaming support
to agent through stream:true in the agent config

- [#1304](browserbase/stagehand#1304)
[`d5e119b`](browserbase/stagehand@d5e119b)
Thanks [@miguelg719](https://github.com/miguelg719)! - Add support for
Microsoft's Fara-7B

- [#1346](browserbase/stagehand#1346)
[`4e051b2`](browserbase/stagehand@4e051b2)
Thanks [@seanmcguire12](https://github.com/seanmcguire12)! - fix: don't
attach to targets twice

- [#1327](browserbase/stagehand#1327)
[`6b5a3c9`](browserbase/stagehand@6b5a3c9)
Thanks [@miguelg719](https://github.com/miguelg719)! - Informed error
parsing from api

- [#1335](browserbase/stagehand#1335)
[`bb85ad9`](browserbase/stagehand@bb85ad9)
Thanks [@seanmcguire12](https://github.com/seanmcguire12)! - add support
for page.addInitScript()

- [#1331](browserbase/stagehand#1331)
[`88d28cc`](browserbase/stagehand@88d28cc)
Thanks [@seanmcguire12](https://github.com/seanmcguire12)! - fix:
page.evaluate() now works with scripts injected via
context.addInitScript()

- [#1316](browserbase/stagehand#1316)
[`45bcef0`](browserbase/stagehand@45bcef0)
Thanks [@tkattkat](https://github.com/tkattkat)! - Add support for
callbacks in stagehand agent

- [#1374](browserbase/stagehand#1374)
[`6aa9d45`](browserbase/stagehand@6aa9d45)
Thanks [@miguelg719](https://github.com/miguelg719)! - Fix key action
mapping in Anthropic CUA

- [#1330](browserbase/stagehand#1330)
[`d382084`](browserbase/stagehand@d382084)
Thanks [@seanmcguire12](https://github.com/seanmcguire12)! - fix: make
act, extract, and observe respect user defined timeout param

- [#1336](browserbase/stagehand#1336)
[`1df08cc`](browserbase/stagehand@1df08cc)
Thanks [@tkattkat](https://github.com/tkattkat)! - Patch agent on api

- [#1358](browserbase/stagehand#1358)
[`2b56600`](browserbase/stagehand@2b56600)
Thanks [@tkattkat](https://github.com/tkattkat)! - Add support for 4.5
opus in cua agent

## @browserbasehq/[email protected]

### Patch Changes

- [#1364](browserbase/stagehand#1364)
[`ca0630e`](browserbase/stagehand@ca0630e)
Thanks [@tkattkat](https://github.com/tkattkat)! - Update model handling
in agent evals cli

- Updated dependencies
\[[`605ed6b`](browserbase/stagehand@605ed6b),
[`34e7e5b`](browserbase/stagehand@34e7e5b),
[`943d2d7`](browserbase/stagehand@943d2d7),
[`0e95cd2`](browserbase/stagehand@0e95cd2),
[`d4237e4`](browserbase/stagehand@d4237e4),
[`86975e7`](browserbase/stagehand@86975e7),
[`d5e119b`](browserbase/stagehand@d5e119b),
[`4e051b2`](browserbase/stagehand@4e051b2),
[`6b5a3c9`](browserbase/stagehand@6b5a3c9),
[`bb85ad9`](browserbase/stagehand@bb85ad9),
[`88d28cc`](browserbase/stagehand@88d28cc),
[`45bcef0`](browserbase/stagehand@45bcef0),
[`6aa9d45`](browserbase/stagehand@6aa9d45),
[`d382084`](browserbase/stagehand@d382084),
[`1df08cc`](browserbase/stagehand@1df08cc),
[`2b56600`](browserbase/stagehand@2b56600)]:
    -   @browserbasehq/[email protected]

## @browserbasehq/[email protected]

### Patch Changes

- Updated dependencies
\[[`605ed6b`](browserbase/stagehand@605ed6b),
[`34e7e5b`](browserbase/stagehand@34e7e5b),
[`943d2d7`](browserbase/stagehand@943d2d7),
[`0e95cd2`](browserbase/stagehand@0e95cd2),
[`d4237e4`](browserbase/stagehand@d4237e4),
[`86975e7`](browserbase/stagehand@86975e7),
[`d5e119b`](browserbase/stagehand@d5e119b),
[`4e051b2`](browserbase/stagehand@4e051b2),
[`6b5a3c9`](browserbase/stagehand@6b5a3c9),
[`bb85ad9`](browserbase/stagehand@bb85ad9),
[`88d28cc`](browserbase/stagehand@88d28cc),
[`45bcef0`](browserbase/stagehand@45bcef0),
[`6aa9d45`](browserbase/stagehand@6aa9d45),
[`d382084`](browserbase/stagehand@d382084),
[`1df08cc`](browserbase/stagehand@1df08cc),
[`2b56600`](browserbase/stagehand@2b56600)]:
    -   @browserbasehq/[email protected]

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants