BrowserTrace: local trace viewer for failed Skyvern task/workflow runs

[aaronlab]

I built BrowserTrace, a small MIT-licensed local trace viewer for failed AI browser-agent runs, and it includes a lightweight Skyvern-shaped wrapper.

Why I am sharing it here: when a Skyvern task or workflow fails, the useful debugging context may be spread across workflow state, browser state, task/run IDs, selected actions, screenshots, model/tool decisions, and the final exception. BrowserTrace records high-level task/workflow calls as a local timeline so failed runs are easier to inspect and share safely.

Minimal shape:

from browsertrace import Tracer
from browsertrace.integrations.skyvern import wrap_skyvern

tracer = Tracer()
skyvern = wrap_skyvern(client, tracer, name="skyvern task run")

await skyvern.run_task("Find the invoice total")
# or: await skyvern.run_workflow(...)

The wrapper keeps Skyvern optional and works with Skyvern-shaped async or sync clients. It records high-level run_task and run_workflow calls, method inputs, returned run IDs/status when available, nested run/status fields, and step-level errors.

Quick no-API trial path from PyPI:

uvx --from "browsertrace[ui]" browsertrace doctor
uvx --from "browsertrace[ui]" browsertrace demo
uvx --from "browsertrace[ui]" browsertrace

Persistent install from PyPI:

pip install "browsertrace[ui]"

Repo: https://github.com/aaronlab/browsertrace
PyPI: https://pypi.org/project/browsertrace/
Skyvern guide: https://aaronlab.github.io/browsertrace/skyvern-debugging.html

Feedback I am looking for from Skyvern users: when a task/workflow fails, which artifacts matter most: workflow state, selected elements/selectors, extracted data, retry/history state, screenshots, model output, task prompts, run IDs/status, or something else?

No star/upvote ask; I am trying to make the Skyvern failure report useful before expanding the adapter.