mirror/browser
1
0
Fork 0
mirror of https://github.com/lightpanda-io/browser.git synced 2026-06-11 01:25:53 -04:00
Code Issues Packages Projects Releases Wiki Activity

tools: add html tool to dump raw page HTML

Browse Source 
Adds a new `html` tool to dump the full raw HTML of the current page.
This is useful for debugging and capturing fixtures.
This commit is contained in:
Adrià Arrufat
2026-05-22 22:31:34 +02:00
parent 056912bbee
commit 2defc0e829
2 changed files with 34 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

24
src/browser/tools.zig
View File

@@ -36,6 +36,7 @@ pub const Tool = enum {
goto,
search,
markdown,
html,
links,
eval,
extract,
@@ -65,7 +66,7 @@ pub const Tool = enum {
pub fn isRecorded(self: Tool) bool {
return switch (self) {
.goto, .eval, .extract, .click, .fill, .scroll, .waitForSelector, .waitForScript, .hover, .press, .selectOption, .setChecked => true,
.search, .markdown, .links, .tree, .nodeDetails, .interactiveElements, .structuredData, .detectForms, .findElement, .consoleLogs, .getUrl, .getCookies, .getEnv => false,
.search, .markdown, .html, .links, .tree, .nodeDetails, .interactiveElements, .structuredData, .detectForms, .findElement, .consoleLogs, .getUrl, .getCookies, .getEnv => false,
};
}
@@ -75,7 +76,7 @@ pub const Tool = enum {
pub fn canHeal(self: Tool) bool {
return switch (self) {
.click, .fill, .scroll, .waitForSelector, .waitForScript, .hover, .press, .selectOption, .setChecked, .extract => true,
.goto, .search, .markdown, .links, .eval, .tree, .nodeDetails, .interactiveElements, .structuredData, .detectForms, .findElement, .consoleLogs, .getUrl, .getCookies, .getEnv => false,
.goto, .search, .markdown, .html, .links, .eval, .tree, .nodeDetails, .interactiveElements, .structuredData, .detectForms, .findElement, .consoleLogs, .getUrl, .getCookies, .getEnv => false,
};
}
@@ -85,7 +86,7 @@ pub const Tool = enum {
pub fn needsLocator(self: Tool) bool {
return switch (self) {
.click, .fill, .hover, .selectOption, .setChecked => true,
.goto, .search, .markdown, .links, .eval, .extract, .tree, .nodeDetails, .interactiveElements, .structuredData, .detectForms, .scroll, .waitForSelector, .waitForScript, .press, .findElement, .consoleLogs, .getUrl, .getCookies, .getEnv => false,
.goto, .search, .markdown, .html, .links, .eval, .extract, .tree, .nodeDetails, .interactiveElements, .structuredData, .detectForms, .scroll, .waitForSelector, .waitForScript, .press, .findElement, .consoleLogs, .getUrl, .getCookies, .getEnv => false,
};
}
@@ -93,7 +94,7 @@ pub const Tool = enum {
/// markdown, eval return value) rather than a status line on stderr.
pub fn producesData(self: Tool) bool {
return switch (self) {
.search, .markdown, .links, .eval, .extract, .tree, .nodeDetails, .interactiveElements, .structuredData, .detectForms, .findElement, .consoleLogs, .getUrl, .getCookies, .getEnv => true,
.search, .markdown, .html, .links, .eval, .extract, .tree, .nodeDetails, .interactiveElements, .structuredData, .detectForms, .findElement, .consoleLogs, .getUrl, .getCookies, .getEnv => true,
.goto, .click, .fill, .scroll, .waitForSelector, .waitForScript, .hover, .press, .selectOption, .setChecked => false,
};
}
@@ -103,7 +104,7 @@ pub const Tool = enum {
pub fn isRetryable(self: Tool) bool {
return switch (self) {
.fill, .setChecked, .selectOption => true,
.goto, .search, .markdown, .links, .eval, .extract, .tree, .nodeDetails, .interactiveElements, .structuredData, .detectForms, .click, .scroll, .waitForSelector, .waitForScript, .hover, .press, .findElement, .consoleLogs, .getUrl, .getCookies, .getEnv => false,
.goto, .search, .markdown, .html, .links, .eval, .extract, .tree, .nodeDetails, .interactiveElements, .structuredData, .detectForms, .click, .scroll, .waitForSelector, .waitForScript, .hover, .press, .findElement, .consoleLogs, .getUrl, .getCookies, .getEnv => false,
};
}
@@ -152,6 +153,10 @@ pub const Tool = enum {
.description = "Get the page content in markdown format. If a url is provided, it navigates to that url first.",
.input_schema = url_params_schema,
},
.html => .{
.description = "Dump the full raw HTML of the current page (doctype + document element). If a url is provided, it navigates to that url first. Prefer `markdown` or `tree` for LLM consumption — `html` is verbose and noisy. Use it for debugging or capturing fixtures.",
.input_schema = url_params_schema,
},
.links => .{
.description = "Extract all links in the opened page. If a url is provided, it navigates to that url first.",
.input_schema = url_params_schema,
@@ -553,6 +558,7 @@ fn dispatch(
.goto => .{ .text = try execGoto(arena, session, registry, substituted) },
.search => .{ .text = try execSearch(arena, session, registry, substituted) },
.markdown => .{ .text = try execMarkdown(arena, session, registry, substituted) },
.html => .{ .text = try execHtml(arena, session, registry, substituted) },
.links => .{ .text = try execLinks(arena, session, registry, substituted) },
.tree => .{ .text = try execTree(arena, session, registry, substituted) },
.nodeDetails => .{ .text = try execNodeDetails(arena, session, registry, substituted) },
@@ -753,6 +759,14 @@ fn execMarkdown(arena: std.mem.Allocator, session: *lp.Session, registry: *CDPNo
return renderFrameMarkdown(arena, page);
}
fn execHtml(arena: std.mem.Allocator, session: *lp.Session, registry: *CDPNode.Registry, arguments: ?std.json.Value) ToolError![]const u8 {
const args = try parseArgsOrDefault(UrlParams, arena, arguments);
const page = try ensurePage(session, registry, args.url, args.timeout, args.waitUntil);
var aw: std.Io.Writer.Allocating = .init(arena);
lp.dump.root(page.document, .{}, &aw.writer, page) catch return ToolError.InternalError;
return aw.written();
}
fn execLinks(arena: std.mem.Allocator, session: *lp.Session, registry: *CDPNode.Registry, arguments: ?std.json.Value) ToolError![]const u8 {
const args = try parseArgsOrDefault(UrlParams, arena, arguments);
const page = try ensurePage(session, registry, args.url, args.timeout, args.waitUntil);

15
src/mcp/tools.zig
View File

@@ -898,6 +898,21 @@ test "MCP - waitForSelector: timeout" {
}, out.written());
}
test "MCP - html dumps doctype + document element" {
defer testing.reset();
var out: std.io.Writer.Allocating = .init(testing.arena_allocator);
const server = try testLoadPage("http://localhost:9582/src/browser/tests/mcp_press_form.html", &out.writer);
defer server.deinit();
const msg =
\\{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"html"}}
;
try router.handleMessage(server, testing.arena_allocator, msg);
try testing.expect(std.mem.indexOf(u8, out.written(), "<!DOCTYPE html>") != null);
try testing.expect(std.mem.indexOf(u8, out.written(), "<form id=\\\"f\\\"") != null);
try testing.expect(std.mem.indexOf(u8, out.written(), "<input id=\\\"q\\\"") != null);
}
test "MCP - waitForScript: truthy returns, falsy times out" {
defer testing.reset();
var out: std.io.Writer.Allocating = .init(testing.arena_allocator);