Strixa AI
TopicsSearchPricing
Sign inStart tracking

Signal Detail

Loading signal detail

Reading this signal card's judgment, risks, and evidence.

Strixa AI
TopicsSearchPricing
Sign inStart tracking
Back to Signal Feed
CodeTracked since May 17, 2026

Screenshot inline base64 blob in browser-use breaks follow-up Claude turns

An open issue reports that when browser-use triggers a CDP screenshot action, the returned raw base64 PNG is persisted in Claude Code conversation context and then resent on each turn, causing every subsequent message to fail with `400 invalid_request_error: Could not process image`. The failure is effectively unrecoverable within the same session and blocks normal interaction until the user opens a new session.

browser-usebrowser-harnessClaude CodeCDP

Signal Intelligence

Confidence Level97%
Worth Continued TrackingFinal score 84% with 1 evidence item.
Primary Impact Areas
browser-usebrowser-harnessClaude Code

What Happened

  • An open issue reports that when browser-use triggers a CDP screenshot action, the returned raw base64 PNG is persisted in Claude Code conversation context and then resent on each turn, causing every subsequent message to fail with `400 invalid_request_error: Could not process image`. The failure is effectively unrecoverable within the same session and blocks normal interaction until the user opens a new session.
  • An open issue reports that when browser-use triggers a CDP screenshot action, the returned raw base64 PNG is persisted in Claude Code conversation context and then resent on each turn, causing every subsequent message to fail with `400 invalid_request_error: Could not process image`. The failure is effectively unrecoverable within the same session and blocks normal interaction until the user opens a new session.
  • 1 evidence item attached for review.

What is Different

Before

Scattered source updates, isolated context, and manual follow-up across multiple feeds.

Now

The issue exposes a concrete integration correctness bug: screenshot tool output is inlined as raw image data in chat/tool history, and that format is rejected by the API in later turns. This pins the fix target to response-shaping behavior (e.g., avoid default inlining, return path/URL, or require explicit screenshot payload requests) rather than generic reliability tuning.

Why Track This

Why It Matters

Users running browser automation through Claude Code can have an active session become unusable after a single screenshot step, so they lose continuity and must start over, which is disruptive for interactive agent workflows. The technical trigger is likely replaying a large or malformed base64 image blob from `Page.captureScreenshot` in every request, which repeatedly trips Anthropic validation; teams should watch whether upcoming changes normalize image handling (media_type, size/compression, and conditional return) before assuming screenshot-based flows are safe at scale.

Impact

Users running browser automation through Claude Code can have an active session become unusable after a single screenshot step, so they lose continuity and must start over, which is disruptive for interactive agent workflows. The technical trigger is likely replaying a large or malformed base64 image blob from `Page.captureScreenshot` in every request, which repeatedly trips Anthropic validation; teams should watch whether upcoming changes normalize image handling (media_type, size/compression, and conditional return) before assuming screenshot-based flows are safe at scale.

What To Watch Next

  • Watch whether browser-use becomes a repeated pattern.
  • Track follow-up changes around AI Agents.
  • Compare future signals against this evidence trail.
  • Re-check risk flags: session_bricks_after_screenshot, raw_image_blob_replayed_in_context.
Open Topic TimelineOpen Technical EventOpen Original Sourcesession_bricks_after_screenshot / raw_image_blob_replayed_in_context / manual_session_restart_required_for_recovery / image_payload_format_or_size_validation_risk

Supporting Evidence

GITHUB ISSUEHigh Trust

browser-use/browser-use Issue #4742: Bug: Screenshot blob in tool result poisons conversation context → API 400 on all subsequent turns

After a screenshot action, every follow-up request returns `API Error: 400 {"type":"error","error":{"type":"invalid_request_error","message":"Could not process image"}}`, and no message can pass through.