mirror of https://github.com/hanzoai/python-sdk.git synced 2026-10-23 20:23:00 +00:00

Files

T

zeekay 450f65932d browser teaches the snapshot loop in its description and the rest on help

The tool put ~120 actions in a Literal enum and ~60 parameters in its
schema, all of it in every agent's context, and most of it Playwright
locator builders that returned a count. Three hand-kept lists (the enum,
extension_actions, _zap_method_for) disagreed: snapshot, wait and
select_option routed to the extension under names it did not answer.

ACTIONS is now the one table: each action once, its topic, its usage
line, and the extension method that serves it. The schema carries the
core loop (navigate, snapshot, click, fill, type, press, read,
screenshot, evaluate, wait, tabs, help) with an open `action` string;
help prints the rest by topic, and their parameters travel in `args`.

snapshot, click-by-ref and friends go to the extension's page engine
(hanzo.snapshot / hanzo.act / hanzo.read / hanzo.wait / hanzo.navigate,
screenshot annotate=true to hanzo.annotate). A tree or a page answers as
text; a refusal ("Stale ref @e4 … run snapshot again", "covered by
<div#consent>") is the error, not a success wrapping "ERR:", and is
never retried on Playwright, which has no refs to act on.

Aliases and no-ops are gone: select_option (select), get_inner_text and
get_value (get_text), clear (fill), content (get_html), new_page
(new_tab), press_key, mouse_wheel (scroll), list_browsers and
list_mcp_instances (browsers), use_browser, set_default_browser,
claim/release_browser, the locator/get_by_*/first/last/nth/filter/all
builders, is_hidden, frame, main_frame, on and off.

Claude-Session: https://claude.ai/code/session_01Dc8DXc3fEQCRHUa6XWK4Ku

2026-09-29 13:05:21 -07:00

1.4 KiB

Raw Permalink Blame History

hanzo-tools-browser

Browser tools for Hanzo MCP: browser drives the user's own browser through the Hanzo extension (headless Playwright when none is connected), playwright is the same tool pinned to Playwright, and cdp sends a raw DevTools method.

Installation

pip install hanzo-tools-browser
pip install 'hanzo-tools-browser[playwright]' && playwright install chromium   # headless fallback

The loop

browser(action="navigate", url="https://example.com")
browser(action="snapshot", interactive=True)       # - button "Sign in" [ref=e2]
browser(action="click", selector="@e2")
browser(action="fill", selector="@e3", text="[email protected]")
browser(action="press", key="Enter")
browser(action="read", outline=True)               # the page as markdown
browser(action="screenshot", annotate=True)        # labels [N] = @eN, with a legend

selector takes a ref from the last snapshot or a CSS selector. A stale ref, or a click on an element covered by a banner or modal, is refused with what to do next.

Progressive surface

The schema carries the core actions: navigate, snapshot, click, fill, type, press, read, screenshot, evaluate, wait, tabs, help. browser(action="help") lists the rest by topic, and their parameters go in args:

browser(action="select", selector="@e4", args={"value": "Weekly"})

License

MIT

1.4 KiB Raw Permalink Blame History