@raultov/browser-control-mcp 0.9.0-beta.13 → 0.10.0-beta.15
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/CHANGELOG.md +100 -0
- package/README.md +24 -14
- package/npm-shrinkwrap.json +2 -2
- package/package.json +2 -2
package/CHANGELOG.md
CHANGED
|
@@ -10,6 +10,106 @@ from a copy of its code: entries describe the differences from that release.
|
|
|
10
10
|
|
|
11
11
|
## [Unreleased]
|
|
12
12
|
|
|
13
|
+
## [0.10.0-beta.15] - 2026-11-02
|
|
14
|
+
|
|
15
|
+
Public **BETA**. Fixes from a comparison round against Playwright MCP,
|
|
16
|
+
chrome-devtools-mcp and a Puppeteer MCP (now repeatable with
|
|
17
|
+
`make compare`): `snapshot` lists a native `<select>`'s options, marks
|
|
18
|
+
rejected fields `[invalid]` with their error, and gives refs to draggable
|
|
19
|
+
elements and drop zones; `fill_input` picks several options of a
|
|
20
|
+
`<select multiple>`; `press_key` takes `Control+a`; `navigate` and `reload`
|
|
21
|
+
name HTTP authentication (Firefox waited 30 s on its prompt). Behaviour
|
|
22
|
+
changes: on a `<select multiple>`, `fill_input` selects exactly the options
|
|
23
|
+
its text names (it added one); the error for a missing option names it.
|
|
24
|
+
|
|
25
|
+
### Fixed
|
|
26
|
+
|
|
27
|
+
- `snapshot` and `find_in_snapshot` list the options of a native `<select>`
|
|
28
|
+
under its combobox, with `[selected]` and `[disabled]` and optgroups as
|
|
29
|
+
groups: only the chosen one showed, so an agent had to guess a label to
|
|
30
|
+
learn the choices. `snapshot` lists the first 25 and says how many more;
|
|
31
|
+
`find_in_snapshot` searches them all. Options have no ref: `fill_input`
|
|
32
|
+
picks one by its label or value. (#46)
|
|
33
|
+
- `navigate` and `reload` to a page behind HTTP authentication (401) fail at
|
|
34
|
+
once saying so, with the scheme and realm of its challenge and how to log
|
|
35
|
+
in (the credentials in the URL, which the browser keeps for the session):
|
|
36
|
+
Chrome only said `net::ERR_INVALID_AUTH_CREDENTIALS`, and Firefox showed
|
|
37
|
+
its prompt while the call waited 30 s and answered the previous page.
|
|
38
|
+
(#50)
|
|
39
|
+
- `press_key` takes a key written with its modifiers, `Control+a`,
|
|
40
|
+
`Shift+Tab` or `ControlOrMeta+c` (Meta on macOS, Control elsewhere), as
|
|
41
|
+
agents write it after Playwright; they add to `modifiers`. It failed with
|
|
42
|
+
`Unknown key 'Control+a'`. (#51)
|
|
43
|
+
- `get_network_logs`' note on responses without a body says they are still
|
|
44
|
+
loading or were never read by the page: Chrome never finishes a `fetch`
|
|
45
|
+
whose body the page did not consume, and the old wording (an open event
|
|
46
|
+
stream, no body until it ends) suggested that waiting would help. (#51)
|
|
47
|
+
- `snapshot` marks a field the page or the browser rejects `[invalid]`
|
|
48
|
+
(`aria-invalid`, or `:user-invalid` once the user changed it or tried to
|
|
49
|
+
submit), with `[error="..."]` from its `aria-errormessage`,
|
|
50
|
+
`aria-describedby` or the browser's validation message: after a submit an
|
|
51
|
+
agent could not tell which fields were rejected. (#49)
|
|
52
|
+
- `snapshot` gives a ref to an element made draggable (`draggable="true"`,
|
|
53
|
+
marked `[draggable]`) and to a drop zone (an `ondrop` or `ondragover`
|
|
54
|
+
attribute): both were plain text, so `drag` needed CSS selectors. (#48)
|
|
55
|
+
- `fill_input` (and `fill_form`) on a `<select multiple>` selects exactly
|
|
56
|
+
the options named one per line of `text` (none for an empty one), firing
|
|
57
|
+
`input` and `change` once; an unknown line fails naming it, with nothing
|
|
58
|
+
changed. It could only add one option at a time. (#47)
|
|
59
|
+
|
|
60
|
+
## [0.10.0-beta.14] - 2026-11-01
|
|
61
|
+
|
|
62
|
+
Public **BETA**. `emulate` gains `user_agent`, `locale`, `timezone` and
|
|
63
|
+
`geolocation` on both browsers, and each call now adds to the earlier ones;
|
|
64
|
+
input actions say when they opened a tab; `evaluate_js` takes top-level
|
|
65
|
+
`await`; Chrome's snapshot reads closed shadow roots. Behaviour changes:
|
|
66
|
+
`navigate` and `reload` fail at once when the page pauses at a breakpoint or
|
|
67
|
+
opens a dialog while loading (Chrome waited 30 s); with a modal `<dialog>`
|
|
68
|
+
open, `snapshot` lists only the dialog; on Firefox a navigation that becomes
|
|
69
|
+
a download fails, as on Chrome; `emulate`'s reply has `user_agent`, `locale`
|
|
70
|
+
and `timezone`.
|
|
71
|
+
|
|
72
|
+
### Fixed
|
|
73
|
+
|
|
74
|
+
- Chrome: `snapshot` and `find_in_snapshot` read closed shadow roots (some
|
|
75
|
+
design systems and third-party widgets), whose elements were invisible:
|
|
76
|
+
Chrome finds them (`DOM.describeNode` with `pierce`) and hands them to
|
|
77
|
+
the outline for that call only, so the page's own scripts never get them;
|
|
78
|
+
their elements are then reached by ref like any other. Pages up to 20 000
|
|
79
|
+
elements; `host >>> inner` selectors still stop at a closed root. Firefox
|
|
80
|
+
has no equivalent: there they stay out of reach. (#42)
|
|
81
|
+
- Firefox: `navigate` to a URL Firefox saves rather than opens (a type such
|
|
82
|
+
as CSV, without `Content-Disposition: attachment`) answered "Navigated
|
|
83
|
+
to" the previous page; it now fails as a download, as on Chrome, from
|
|
84
|
+
Firefox's `browsingContext.downloadWillBegin`. (#39)
|
|
85
|
+
- `evaluate_js` takes `await` at the top level (`const r = await
|
|
86
|
+
fetch(u); r.json()`), answering the last statement's value as a console
|
|
87
|
+
does: it failed with a SyntaxError on both browsers. A symbol is described
|
|
88
|
+
as `Symbol(x)` on both (Firefox answered an empty object), and a function
|
|
89
|
+
on Chrome by its code (it answered `{}`). (#43)
|
|
90
|
+
- `emulate`: each call adds to the earlier ones, for every option (Chrome
|
|
91
|
+
kept an earlier viewport but dropped the earlier media settings), and it
|
|
92
|
+
emulates `user_agent`, `locale`, `timezone` and `geolocation` on both
|
|
93
|
+
browsers (the page is granted the geolocation permission, and the reply
|
|
94
|
+
says the position it reads). An unknown time zone or locale fails without
|
|
95
|
+
changing anything (Chrome kept the old zone silently). (#43)
|
|
96
|
+
- `click_element`, `press_key` and the other input tools say when the
|
|
97
|
+
action opened tabs (a `target=_blank` link, `window.open`), naming each
|
|
98
|
+
with its URL: the new tab was registered, but nothing told the agent to
|
|
99
|
+
look at it. (#41)
|
|
100
|
+
- `snapshot` and `find_in_snapshot` show `<progress>` and `<meter>`
|
|
101
|
+
(`progressbar`, `meter`, with their value as `40/100`), say whether a
|
|
102
|
+
`<details>` summary is expanded, mark an open modal `<dialog>` `[modal]`,
|
|
103
|
+
and leave out what the page made inert: `inert` subtrees and, while a
|
|
104
|
+
modal dialog is open, everything outside it (it was listed, and clicking
|
|
105
|
+
it failed as covered). (#40)
|
|
106
|
+
- Chrome: `navigate` and `reload` to a page that pauses at a breakpoint
|
|
107
|
+
while it loads (in the page or in a frame of another origin) waited the
|
|
108
|
+
whole 30 s, then answered as if the page were slow. They now fail at once
|
|
109
|
+
naming where it paused, as on Firefox; `resume` lets the page finish
|
|
110
|
+
loading. A page opening a dialog while it loads (`alert()` in an inline
|
|
111
|
+
script) fails at once too, pointing to `handle_dialog`. (#45)
|
|
112
|
+
|
|
13
113
|
## [0.9.0-beta.13] - 2026-11-01
|
|
14
114
|
|
|
15
115
|
Public **BETA**. Frames of another origin are now observed and debugged, not
|
package/README.md
CHANGED
|
@@ -13,7 +13,7 @@ It is the successor of
|
|
|
13
13
|
from a copy of its code) and complements Playwright MCP or
|
|
14
14
|
chrome-devtools-mcp with runtime JavaScript debugging on both browsers.
|
|
15
15
|
|
|
16
|
-
> **Status: BETA (0.
|
|
16
|
+
> **Status: BETA (0.10.0-beta.15).** This version contains known bugs and differences
|
|
17
17
|
> between Chrome and Firefox (see [Limitations and TODO](#limitations-and-todo))
|
|
18
18
|
> and parts that have not been verified (see [Untested](#untested)). Tool
|
|
19
19
|
> replies may change in incompatible ways before 1.0. Please report problems
|
|
@@ -27,7 +27,9 @@ Measured on 2026-10-31 with browser-control-mcp built from the current
|
|
|
27
27
|
sources, `@playwright/mcp` 0.0.83 and `chrome-devtools-mcp` 1.12.0, driven by
|
|
28
28
|
the same stdio client through the same tasks on Chrome 156 (headless,
|
|
29
29
|
Linux). Every outcome was checked in the page, not taken from the reply.
|
|
30
|
-
Tokens are estimated as characters / 4 for every server.
|
|
30
|
+
Tokens are estimated as characters / 4 for every server. browser-control's
|
|
31
|
+
tool lists and the HTTP authentication row were measured again on
|
|
32
|
+
2026-11-02, after the fixes of #46 to #51.
|
|
31
33
|
|
|
32
34
|
**What each one can do**
|
|
33
35
|
|
|
@@ -41,6 +43,7 @@ Tokens are estimated as characters / 4 for every server.
|
|
|
41
43
|
| Snapshot of a long page | **bounded at 50 000 characters**, `find_in_snapshot` searches the rest | unbounded | unbounded |
|
|
42
44
|
| Waits for the page to react after an action | ✓ ~50 ms when nothing reacts | ✓ ~500 ms | ✓ ~200 ms |
|
|
43
45
|
| Clicking a covered or hidden element | **fails at once, naming the overlay** | fails after a 5 s timeout | reports success (the overlay gets the click) |
|
|
46
|
+
| Page behind HTTP authentication (401) | **fails at once naming its scheme and realm and how to log in** (credentials in the URL), on Chrome and Firefox | fails with `net::ERR_INVALID_AUTH_CREDENTIALS` | reports success, with `net::ERR_INVALID_AUTH_CREDENTIALS` in the text |
|
|
44
47
|
| Mock requests, offline mode | ✓ | ✓ | offline only |
|
|
45
48
|
| PDF | ✓ file or inline | ✓ file | ✗ |
|
|
46
49
|
| Cookies and web storage | ✓ values hidden unless asked | ✓ values shown | ✗ |
|
|
@@ -55,8 +58,8 @@ Tokens are estimated as characters / 4 for every server.
|
|
|
55
58
|
|
|
56
59
|
| Task | browser-control-mcp | Playwright MCP | chrome-devtools-mcp |
|
|
57
60
|
| --- | --- | --- | --- |
|
|
58
|
-
| Tool list in context, default tools | 6
|
|
59
|
-
| Tool list in context, every opt-in tool on |
|
|
61
|
+
| Tool list in context, default tools | 6 730 tokens | 5 330 tokens | 7 770 tokens |
|
|
62
|
+
| Tool list in context, every opt-in tool on | 10 390 tokens | 12 040 tokens | not measured (optional categories) |
|
|
60
63
|
| Snapshot of Hacker News | **4 588 tokens** | 12 356 tokens | 6 869 tokens |
|
|
61
64
|
| Snapshot of Wikipedia "Web browser" | **12 643 tokens** (cut, searchable) | 34 349 tokens | 27 656 tokens |
|
|
62
65
|
| Snapshot of MDN `<input>` | **12 625 tokens** (cut, searchable) | 158 077 tokens | 76 654 tokens |
|
|
@@ -119,29 +122,29 @@ see the note.
|
|
|
119
122
|
| `close_tab` | ✓ | ✓ | |
|
|
120
123
|
| `list_tabs` | ✓ | ✓ | |
|
|
121
124
|
| `switch_tab` | ✓ | ✓ | |
|
|
122
|
-
| `navigate` | ✓ | ✓ | `history: "back"` or `"forward"` instead of `url` moves one entry in the tab's history and waits for the page to load (also when it comes from the back-forward cache, which fires no `load`); at either end it fails without moving |
|
|
123
|
-
| `reload` | ✓ | ✓ | |
|
|
125
|
+
| `navigate` | ✓ | ✓ | `history: "back"` or `"forward"` instead of `url` moves one entry in the tab's history and waits for the page to load (also when it comes from the back-forward cache, which fires no `load`); at either end it fails without moving; a URL the browser saves instead of opening (a download: `Content-Disposition: attachment`, or a type such as CSV) fails saying so, and the tab stays where it was; a page behind HTTP authentication (401) fails at once naming its scheme and realm and how to log in (credentials in the URL, `http://user:password@host/path`, which the browser keeps for the session); a page that pauses at a breakpoint or opens a dialog while loading fails at once naming it (see `resume`, `handle_dialog`) |
|
|
126
|
+
| `reload` | ✓ | ✓ | a pause, a dialog or HTTP authentication while loading fails at once, as with `navigate` |
|
|
124
127
|
| `scroll` | ✓ | ✓ | reports the window positions before and after |
|
|
125
|
-
| `emulate` | ✓ | partial | `emulation` group; `width`/`height`, `device_scale_factor`, `mobile` (touch), `color_scheme`, `reduced_motion`, `forced_colors`, `media` (`print`), `reset`; answers what the page now sees; Firefox:
|
|
128
|
+
| `emulate` | ✓ | partial | `emulation` group; `width`/`height`, `device_scale_factor`, `mobile` (touch), `color_scheme`, `reduced_motion`, `forced_colors`, `media` (`print`), `user_agent`, `locale`, `timezone` (checked by the page's `Intl` first), `geolocation` (the page is granted the permission to read it), `reset`; each call adds to the earlier ones; answers what the page now sees (with the position it reads); Firefox: no `mobile` or media options (BiDi has no media or touch emulation yet; they fail naming them) |
|
|
126
129
|
| `wait_for` | ✓ | ✓ | waits (checking every 100 ms, at most 30 s) until a visible text appears or is gone, a selector is `visible` / `hidden` / `attached` / `detached`, or the URL contains a text; a dialog or debugger pause ends the wait at once |
|
|
127
130
|
| `inspect_dom` | ✓ | ✓ | includes open shadow roots and same-origin iframes (cross-origin ones are noted); at most 100 000 characters without `query` (cut in the page) |
|
|
128
|
-
| `snapshot` | ✓ | ✓ | outline of the page: role, name, states, value and a ref (`e12`) per element, through open shadow roots and same-origin iframes; frames of another origin are read too (at most 20, 5 deep), their iframe marked `[frame=f2]` and their refs prefixed (`f2e5`), and `selector: "f2"` outlines one; tables without headers (`th`, `thead`, `caption`) are layout and leave only their content, and a cell is named by its text only when it holds nothing else; comboboxes, spinbuttons and sliders built with ARIA show their value (the field's text, `aria-valuetext` or `aria-valuenow`, the option a button-like combobox shows); an element that is only clickable (a handler or a pointer cursor) is one named line
|
|
131
|
+
| `snapshot` | ✓ | ✓ | outline of the page: role, name, states, value and a ref (`e12`) per element, through open shadow roots and same-origin iframes, and on Chrome closed shadow roots too (pages up to 20 000 elements; their elements are reached by ref, not by `>>>`); frames of another origin are read too (at most 20, 5 deep), their iframe marked `[frame=f2]` and their refs prefixed (`f2e5`), and `selector: "f2"` outlines one; tables without headers (`th`, `thead`, `caption`) are layout and leave only their content, and a cell is named by its text only when it holds nothing else; comboboxes, spinbuttons and sliders built with ARIA show their value (the field's text, `aria-valuetext` or `aria-valuenow`, the option a button-like combobox shows), progress bars and meters theirs (`40/100`), and a `<details>` summary `[expanded=true|false]`; a field the page rejects (`aria-invalid`) or the browser does (`:user-invalid`, once the user changed it or tried to submit) is `[invalid]`, with `[error="..."]` from its `aria-errormessage`, `aria-describedby` or the browser's validation message; a native `<select>` lists its options under its combobox (`[selected]`, `[disabled]`, optgroups as groups; no refs: `fill_input` picks one), the first 25 and how many more, all of them for `find_in_snapshot`; what the page made inert is left out: `inert` subtrees and, while a modal `<dialog>` is open (marked `[modal]`), everything outside it; an element that is only clickable (a handler or a pointer cursor), draggable (`draggable="true"`, marked `[draggable]`) or a drop zone (`ondrop`/`ondragover`) is one named line with a ref for `drag`, unless it holds links, buttons or fields, which are listed under it; at most 50 000 characters (`selector` and `depth` narrow it); the same JavaScript on both browsers |
|
|
129
132
|
| `find_in_snapshot` | ✓ | ✓ | `text` (case-insensitive) or `regex` over the whole outline; matching lines under their ancestors, at most 20 |
|
|
130
133
|
| `save_pdf` | ✓ | ✓ | `files` group; prints the tab as the browser's print does (`landscape`, `print_background`, `scale` 0.1–2, `page_ranges` like `1-3, 5`); with `path` (an absolute `.pdf` under the upload roots) writes the file and answers `{path, bytes, pages}`, else returns the PDF as an embedded `application/pdf` resource (at most 5 MB); Chrome `Page.printToPDF`, Firefox `browsingContext.print` |
|
|
131
134
|
| `capture_screenshot` | ✓ | ✓ | `format` `png`, `jpeg` (or `jpg`) or `webp`; `full_page` at most 16 384 px per side (a text note says when cut); `selector` captures one element (CSS, `host >>> inner` or a snapshot ref), scrolled into view, whole even when taller than the viewport |
|
|
132
135
|
| `click_element` | ✓ | ✓ | scrolls the element into view; `button` (left/right/middle), `click_count` (2 = double click) and `modifiers`; fails without clicking when it is hidden, covered (its own label on top does not count: styled checkboxes and switches are clicked through it; a covered element first gets the mouse, so controls shown only on hover are revealed and clicked) or a disabled control, whose handler would never run; an element a CSS selector does not find yet, or a disabled control, is checked again for up to 2 s, as pages add elements and enable buttons a moment after the previous action (a ref that names nothing, a hidden or a covered element fail at once); `host >>> inner` selectors reach open shadow roots and same-origin iframes (also in `fill_input`, `scroll`, `wait_for`); a dialog or debugger pause the click causes is noted in a successful reply (`{dialog}` or `{paused, location}`) |
|
|
133
|
-
| `fill_input` | ✓ | ✓ | picks `<select>` options; sets `date`, `time`, `datetime-local`, `month`, `week`, `range` and `color` inputs to the text in their own format (`2026-10-31`, `13:45`, `72`, `#ff0000`), firing `input` and `change`, noting a value the input adjusted (a range snaps to its step) and failing, without changing it, on a malformed one; fails without typing on elements that cannot take the text; `append: true` types after the content instead of replacing it (a number or email field, which has no caret, is retyped whole); an editor of its own (an element with an `EditContext` or a focusable `textbox` role, or the hidden input of a code editor such as Monaco or Ace) gets the keys a user presses: a click on the editor when its input has no visible box, Control+A (Command+A on macOS) or Control+End, then the text, and the reply says its content could not be checked; in a contenteditable element (including rich text editors such as Slate, Lexical, ProseMirror or CodeMirror 6) it selects the content (a range, then Control+A as a user would) and types over it, so the editor records the edit; types key by key on both (key handlers run; newlines and tabs inserted as text), about 1 ms per character; `change` on leaving the field, or at the end when the window has no focus |
|
|
134
|
-
| `press_key` | ✓ | ✓ | one key (a character, Enter, Tab, Escape, arrows, F1–F12...) with Control / Shift / Alt / Meta held, in `selector` (focused first) or the focused element; a shortcut types nothing |
|
|
136
|
+
| `fill_input` | ✓ | ✓ | picks `<select>` options (on a `<select multiple>`, exactly the ones named one per line; an unknown one changes nothing); sets `date`, `time`, `datetime-local`, `month`, `week`, `range` and `color` inputs to the text in their own format (`2026-10-31`, `13:45`, `72`, `#ff0000`), firing `input` and `change`, noting a value the input adjusted (a range snaps to its step) and failing, without changing it, on a malformed one; fails without typing on elements that cannot take the text; `append: true` types after the content instead of replacing it (a number or email field, which has no caret, is retyped whole); an editor of its own (an element with an `EditContext` or a focusable `textbox` role, or the hidden input of a code editor such as Monaco or Ace) gets the keys a user presses: a click on the editor when its input has no visible box, Control+A (Command+A on macOS) or Control+End, then the text, and the reply says its content could not be checked; in a contenteditable element (including rich text editors such as Slate, Lexical, ProseMirror or CodeMirror 6) it selects the content (a range, then Control+A as a user would) and types over it, so the editor records the edit; types key by key on both (key handlers run; newlines and tabs inserted as text), about 1 ms per character; `change` on leaving the field, or at the end when the window has no focus |
|
|
137
|
+
| `press_key` | ✓ | ✓ | one key (a character, Enter, Tab, Escape, arrows, F1–F12...) with Control / Shift / Alt / Meta held (in `modifiers`, or written before the key as `Control+a`, `Shift+Tab` or `ControlOrMeta+c`: Meta on macOS, Control elsewhere), in `selector` (focused first) or the focused element; a shortcut types nothing |
|
|
135
138
|
| `hover` | ✓ | ✓ | moves the mouse over the element (scrolled into view; fails when hidden or covered) |
|
|
136
139
|
| `drag` | ✓ | ✓ | presses on `from` (checked like `click_element`), moves in steps to the centre of `to` (which must then be in the viewport) and releases: sortable lists, boards, sliders; HTML5 drag and drop is real on Chrome and simulated with script events on Firefox (its WebDriver input starts a native drag but never drops it), and the reply says so |
|
|
137
140
|
| `fill_form` | ✓ | ✓ | several fields in one call: text and `<select>` as `fill_input`, a checkbox or radio takes `true`/`false` and is clicked only when its state differs; stops at the first failing field naming what was done; never echoes values |
|
|
138
141
|
| `upload_file` | ✓ | ✓ | `files` group; sets the files of an `<input type=file>` (Chrome `DOM.setFileInputFiles`, Firefox `input.setFiles`); files must be under the upload roots (`--upload-root`, default the working and temp directories) |
|
|
139
142
|
| `handle_dialog` | ✓ | ✓ | accepts or dismisses the open `alert` / `confirm` / `prompt` / `beforeunload` dialog |
|
|
140
|
-
| `evaluate_js` | ✓ | ✓ | `value` at most 100 000 characters (`truncated: true`); Chrome stops a script after 8 s, Firefox cannot (a looping tab is closed with `close_tab`) |
|
|
143
|
+
| `evaluate_js` | ✓ | ✓ | `await` at the top level (`const r = await fetch(u); r.json()`), as in a console; `value` at most 100 000 characters (`truncated: true`); symbols and functions are described; Chrome stops a script after 8 s, Firefox cannot (a looping tab is closed with `close_tab`) |
|
|
141
144
|
| `cookies` | ✓ | ✓ | `storage` group; `get` / `set` / `delete`: the cookies sent to the page (or `url`) or of a `domain`; values hidden unless `show_values: true` (Chrome `Network.*Cookie*`, Firefox BiDi `storage.*` in the tab's partition) |
|
|
142
145
|
| `web_storage` | ✓ | ✓ | `storage` group; `get` / `set` / `delete` / `clear` on the page's `localStorage` or `sessionStorage`; values hidden unless `show_values: true` |
|
|
143
146
|
| `get_console_logs` | ✓ | ✓ | one line per message, oldest first: `[level] text @ url:line` (lines from 1), objects and arrays previewed as the console shows them (`{n: 1}`, `[1, 2, 3]`), repeats in a row folded (`[x3]`), uncaught errors marked with up to 5 stack lines indented; `include_details: true` gives the JSON with every field; a message keeps at most 10 000 characters; Firefox: page errors, CSS warnings and platform messages need the debugger port |
|
|
144
|
-
| `get_network_logs` | ✓ | ✓ | one line per request, oldest first (`id METHOD url => status [type]`, or `failed: <reason>` / `pending`), then one per WebSocket frame (payload preview); `request_id` shows one request's headers and body, `include_details: true` every request's as JSON (a binary body has `response_body_encoding: "base64"`; bodies at most 64 000 characters, none over 2 MB; a WebSocket payload keeps 64 000, `payload_len` is the whole); the browser's own requests (`chrome://`, `devtools://`, `data:`, `about:`, extensions) are not logged; Firefox: WebSocket frames need the debugger port |
|
|
147
|
+
| `get_network_logs` | ✓ | ✓ | one line per request, oldest first (`id METHOD url => status [type]`, or `failed: <reason>` / `pending`), then one per WebSocket frame (payload preview); `request_id` shows one request's headers and body, `include_details: true` every request's as JSON (a binary body has `response_body_encoding: "base64"`; bodies at most 64 000 characters, none over 2 MB; a WebSocket payload keeps 64 000, `payload_len` is the whole); the browser's own requests (`chrome://`, `devtools://`, `data:`, `about:`, extensions) are not logged; a note counts the responses with no body because they are still loading or the page never read them (an open stream; on Chrome, a `fetch` whose body the page did not consume); Firefox: WebSocket frames need the debugger port |
|
|
145
148
|
| `mock_route` | ✓ | ✓ | `network` group; requests whose URL matches a glob (`*`, `?`, `\` escapes) get a mocked `status`, `body` (text, at most 1 MB), `content_type` and `headers` without reaching the network; the latest matching route wins; at most 50 routes per tab, kept until `remove_route` or the tab closes (Chrome `Fetch`, not while `enable_proxy_auth` is active; Firefox a `network` intercept) |
|
|
146
149
|
| `list_routes` | ✓ | ✓ | `network` group; the tab's routes with the `hits` each answered |
|
|
147
150
|
| `remove_route` | ✓ | ✓ | `network` group; one route by id, or all |
|
|
@@ -173,7 +176,9 @@ action nothing reacts to costs those 50 ms and keeps its reply; otherwise
|
|
|
173
176
|
the reply adds `The page reacted (the DOM changed, 1 request(s)) and settled
|
|
174
177
|
after 134 ms.`, or says the page was still changing after 1 s. A click that
|
|
175
178
|
navigates waits for the new page's requests. Use `wait_for` for anything
|
|
176
|
-
slower (a timer, a long request).
|
|
179
|
+
slower (a timer, a long request). An action that opened tabs (a
|
|
180
|
+
`target=_blank` link, `window.open`) says so: `The action opened tab tab-3
|
|
181
|
+
(https://…): call switch_tab with its tab_id to act in it.`
|
|
177
182
|
|
|
178
183
|
The JavaScript debugger tools on Firefox need its debugger (RDP) port: on by
|
|
179
184
|
default for a launched Firefox, `--firefox-rdp-port` for an attached one. The
|
|
@@ -681,12 +686,17 @@ incompatible:
|
|
|
681
686
|
make check # fmt, clippy -D warnings, tests, docs, duplication, packaging, version
|
|
682
687
|
make e2e # E2E smoke against real Chrome and Firefox, headless (E2E_VISIBLE=1 shows them; Linux, macOS)
|
|
683
688
|
make check-all # both
|
|
689
|
+
make compare # the same tasks on this and other browser MCP servers (tests/compare)
|
|
684
690
|
```
|
|
685
691
|
|
|
686
692
|
`make check` is hermetic: it never starts a browser or touches the network.
|
|
687
693
|
`make e2e` runs `tests/run_all_e2e.sh`: the launched-mode smoke for both
|
|
688
694
|
browsers, then the attach mode against browsers started by `tests/chrome.sh`
|
|
689
|
-
and `tests/firefox.sh`.
|
|
695
|
+
and `tests/firefox.sh`. `make compare` runs the scenarios of
|
|
696
|
+
[tests/compare](tests/compare/README.md) (from common tasks to rare ones, each
|
|
697
|
+
with a check) on browser-control and on Playwright MCP, chrome-devtools-mcp
|
|
698
|
+
and a Puppeteer MCP, and compares the outcome, calls and tokens with a saved
|
|
699
|
+
baseline. See [AGENTS.md](AGENTS.md) for the conventions.
|
|
690
700
|
|
|
691
701
|
## Releasing
|
|
692
702
|
|
package/npm-shrinkwrap.json
CHANGED
|
@@ -19,7 +19,7 @@
|
|
|
19
19
|
"hasInstallScript": true,
|
|
20
20
|
"license": "MIT",
|
|
21
21
|
"name": "@raultov/browser-control-mcp",
|
|
22
|
-
"version": "0.
|
|
22
|
+
"version": "0.10.0-beta.15"
|
|
23
23
|
},
|
|
24
24
|
"node_modules/detect-libc": {
|
|
25
25
|
"engines": {
|
|
@@ -48,5 +48,5 @@
|
|
|
48
48
|
}
|
|
49
49
|
},
|
|
50
50
|
"requires": true,
|
|
51
|
-
"version": "0.
|
|
51
|
+
"version": "0.10.0-beta.15"
|
|
52
52
|
}
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"artifactDownloadUrls": [
|
|
3
|
-
"https://github.com/raultov/browser-control-mcp/releases/download/v0.
|
|
3
|
+
"https://github.com/raultov/browser-control-mcp/releases/download/v0.10.0-beta.15"
|
|
4
4
|
],
|
|
5
5
|
"author": "raultov <raultov@protonmail.com>",
|
|
6
6
|
"bin": {
|
|
@@ -92,7 +92,7 @@
|
|
|
92
92
|
"zipExt": ".tar.xz"
|
|
93
93
|
}
|
|
94
94
|
},
|
|
95
|
-
"version": "0.
|
|
95
|
+
"version": "0.10.0-beta.15",
|
|
96
96
|
"volta": {
|
|
97
97
|
"node": "18.14.1",
|
|
98
98
|
"npm": "9.5.0"
|