Open-source Python library for AI browser agents, plus Browser Use Cloud: stealth browsers and a hosted agent API.
Python ai-agentsai-toolsbrowser-agentbrowser-automationbrowser-usecdpcloud-browsercomputer-useheadless-browserllmmcpplaywrightpythonremote-browserstealth-browserweb-automationweb-scraping
Record batch fetch eval gate
Magnus Müller2fda56661e4c
diff --git a/RUST_AGENT_INTEGRATION_FEATURES.md b/RUST_AGENT_INTEGRATION_FEATURES.mdindex ec7dcfd26..158d5a13d 100644--- a/RUST_AGENT_INTEGRATION_FEATURES.md+++ b/RUST_AGENT_INTEGRATION_FEATURES.md@@ -1575,6 +1575,14 @@ Terminal core branch: `magnus/browser-use-rust-main-integration` at terminal mai## Not Verified Yet+- 2026-06-04 post-batch-fetch-helper five-task real_v8 Agent SDK gate:+ - Run `kh7awdj58f4hnkv4dke9237q2h880b5z` used `BU_RUNTIME=brust`, terminal commit `21a180a`, `--browser browser-use-cloud`, `BROWSER_USE_CLOUD_PROXY_COUNTRY_CODE=us`, `BU_BROWSER_SCRIPT_INITIAL_WAIT_MS=7000`, `AGENT_SDK_JUDGE_TIMEOUT_SECONDS=25`, `AGENT_SDK_JUDGE_FALLBACK_ATTEMPTS=3`, fallback timeout default floor `45`, `AGENT_SDK_JUDGE_CONCURRENCY=16`, `LLM_BROWSER_CAPTURE_FPS=0`, `BU_DISABLE_FALLBACK_CAPTURE_GIF=1`, `BU_BROWSER_SCRIPT_SESSION_OUTPUTS=1`, `--use-vision false`, `--judge-type agent_sdk`, `--model claude-sonnet-4-6`, `--eval-model claude-sonnet-4-6`, `--parallel-runs 5`, and `--max-steps 100` for real_v8 tasks 6-10.+ - Browser Use cloud startup was healthy: all five sessions were created around 2026-06-04T11:15:02Z, all five tasks entered Rust-core agent execution, and no CAPTCHA/local-browser misconfiguration appeared.+ - Saved results before the 10-minute cap: task 9 score `0.8` in 266.62s agent time/339.23s total pipeline, task 10 score `0.85` in 411.98s/489.87s, and task 6 score `1.0` in 423.63s/490.24s.+ - Task 9 showed a concrete cost/step improvement over the previous summary-first gate: 17 steps and `725178` total tokens instead of 22 steps and `1267194` total tokens. Task 6 also completed and saved before the cap after being unsaved in the previous gate.+ - Current fresh artifacts did not show direct calls to `http_get_many`, `browser_fetch`, or `browser_fetch_many`, so this gate should not be treated as proof that the new helper API has been adopted by the model in live evals yet.+ - Tasks 7 and 8 were still in agent-stage execution when the cap fired. All five Browser Use cloud sessions were explicitly stopped after the cap, and no eval/terminal subprocesses remained.+ - This gate does not justify scaling to 50: saved-task mean score was about `0.88`, but missing-task-as-zero mean score was `0.53` with only 3/5 tasks saved within the iteration budget. The remaining bottleneck is still long-tail document/research execution and cumulative prompt growth.- 2026-06-04 post-summary-first-browser-script five-task real_v8 Agent SDK gate:- Run `kh77e9y1cjeyygdnvex85qwcrs880hr0` used `BU_RUNTIME=brust`, terminal commit `297930a`, `--browser browser-use-cloud`, `BROWSER_USE_CLOUD_PROXY_COUNTRY_CODE=us`, `BU_BROWSER_SCRIPT_INITIAL_WAIT_MS=7000`, `AGENT_SDK_JUDGE_TIMEOUT_SECONDS=25`, `AGENT_SDK_JUDGE_FALLBACK_ATTEMPTS=3`, fallback timeout default floor `45`, `AGENT_SDK_JUDGE_CONCURRENCY=16`, `LLM_BROWSER_CAPTURE_FPS=0`, `BU_DISABLE_FALLBACK_CAPTURE_GIF=1`, `BU_BROWSER_SCRIPT_SESSION_OUTPUTS=1`, `--use-vision false`, `--judge-type agent_sdk`, `--model claude-sonnet-4-6`, `--eval-model claude-sonnet-4-6`, `--parallel-runs 5`, and `--max-steps 100` for real_v8 tasks 6-10.- Browser Use cloud startup was healthy: all five sessions were created around 2026-06-04T10:54:10Z, all five tasks entered Rust-core agent execution, and no CAPTCHA/local-browser misconfiguration appeared.