feat: add --cookie-file support for JSON cookie files

- --cookie-file reads Chrome extension format ([{name, value, domain, ...}])
- Works with EditThisCookie, Cookie-Editor, and similar browser extensions
- Merges with --cookie when both provided
- MCP scrape tool now accepts cookies parameter
- Closes #7

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Cargo.lock

@@ -3072,7 +3072,7 @@ dependencies = [
 
 [[package]]
 name = "webclaw-cli"
-version = "0.3.0"
+version = "0.3.1"
 dependencies = [
  "clap",
  "dotenvy",
@@ -3092,7 +3092,7 @@ dependencies = [
 
 [[package]]
 name = "webclaw-core"
-version = "0.3.0"
+version = "0.3.1"
 dependencies = [
  "ego-tree",
  "once_cell",
@@ -3110,7 +3110,7 @@ dependencies = [
 
 [[package]]
 name = "webclaw-fetch"
-version = "0.3.0"
+version = "0.3.1"
 dependencies = [
  "calamine",
  "http",
@@ -3148,7 +3148,7 @@ dependencies = [
 
 [[package]]
 name = "webclaw-llm"
-version = "0.3.0"
+version = "0.3.1"
 dependencies = [
  "async-trait",
  "reqwest 0.12.28",
@@ -3161,7 +3161,7 @@ dependencies = [
 
 [[package]]
 name = "webclaw-mcp"
-version = "0.3.0"
+version = "0.3.1"
 dependencies = [
  "dotenvy",
  "reqwest 0.12.28",
@@ -3181,7 +3181,7 @@ dependencies = [
 
 [[package]]
 name = "webclaw-pdf"
-version = "0.3.0"
+version = "0.3.1"
 dependencies = [
  "pdf-extract",
  "thiserror",

crates/webclaw-cli/src/main.rs

@@ -151,6 +151,10 @@ struct Cli {
     #[arg(long)]
     cookie: Option<String>,
 
+    /// JSON cookie file (Chrome extension format: [{name, value, domain, ...}])
+    #[arg(long)]
+    cookie_file: Option<String>,
+
     /// Enable verbose logging
     #[arg(short, long)]
     verbose: bool,
@@ -371,6 +375,24 @@ fn build_fetch_config(cli: &Cli) -> FetchConfig {
         headers.insert("Cookie".to_string(), cookie.clone());
     }
 
+    // --cookie-file: parse JSON array of {name, value, domain, ...}
+    if let Some(ref path) = cli.cookie_file {
+        match parse_cookie_file(path) {
+            Ok(cookie_str) => {
+                // Merge with existing cookies if --cookie was also provided
+                if let Some(existing) = headers.get("Cookie") {
+                    headers.insert("Cookie".to_string(), format!("{existing}; {cookie_str}"));
+                } else {
+                    headers.insert("Cookie".to_string(), cookie_str);
+                }
+            }
+            Err(e) => {
+                eprintln!("error: failed to parse cookie file: {e}");
+                process::exit(1);
+            }
+        }
+    }
+
     FetchConfig {
         browser: cli.browser.clone().into(),
         proxy,
@@ -382,6 +404,29 @@ fn build_fetch_config(cli: &Cli) -> FetchConfig {
     }
 }
 
+/// Parse a JSON cookie file (Chrome extension format) into a Cookie header string.
+/// Supports: [{name, value, domain, path, secure, httpOnly, expirationDate, ...}]
+fn parse_cookie_file(path: &str) -> Result<String, String> {
+    let content = std::fs::read_to_string(path).map_err(|e| format!("cannot read {path}: {e}"))?;
+    let cookies: Vec<serde_json::Value> =
+        serde_json::from_str(&content).map_err(|e| format!("invalid JSON: {e}"))?;
+
+    let pairs: Vec<String> = cookies
+        .iter()
+        .filter_map(|c| {
+            let name = c.get("name")?.as_str()?;
+            let value = c.get("value")?.as_str()?;
+            Some(format!("{name}={value}"))
+        })
+        .collect();
+
+    if pairs.is_empty() {
+        return Err("no cookies found in file".to_string());
+    }
+
+    Ok(pairs.join("; "))
+}
+
 fn build_extraction_options(cli: &Cli) -> ExtractionOptions {
     ExtractionOptions {
         include_selectors: cli

crates/webclaw-mcp/src/server.rs

@@ -139,19 +139,33 @@ impl WebclawMcp {
         let exclude = params.exclude_selectors.unwrap_or_default();
         let main_only = params.only_main_content.unwrap_or(false);
 
-        // Use a custom client if a non-default browser is requested
+        // Build cookie header from params
+        let cookie_header = params
+            .cookies
+            .as_ref()
+            .filter(|c| !c.is_empty())
+            .map(|c| c.join("; "));
+
+        // Use a custom client if non-default browser or cookies are provided
         let is_default_browser = matches!(browser, webclaw_fetch::BrowserProfile::Chrome);
+        let needs_custom = !is_default_browser || cookie_header.is_some();
         let custom_client;
-        let client: &webclaw_fetch::FetchClient = if is_default_browser {
-            &self.fetch_client
-        } else {
+        let client: &webclaw_fetch::FetchClient = if needs_custom {
+            let mut headers = std::collections::HashMap::new();
+            headers.insert("Accept-Language".to_string(), "en-US,en;q=0.9".to_string());
+            if let Some(ref cookies) = cookie_header {
+                headers.insert("Cookie".to_string(), cookies.clone());
+            }
             let config = webclaw_fetch::FetchConfig {
                 browser,
+                headers,
                 ..Default::default()
             };
             custom_client = webclaw_fetch::FetchClient::new(config)
                 .map_err(|e| format!("Failed to build client: {e}"))?;
             &custom_client
+        } else {
+            &self.fetch_client
         };
 
         let formats = [format];

crates/webclaw-mcp/src/tools.rs

@@ -18,6 +18,8 @@ pub struct ScrapeParams {
     pub only_main_content: Option<bool>,
     /// Browser profile: "chrome" (default), "firefox", or "random"
     pub browser: Option<String>,
+    /// Cookies to send with the request (e.g. ["name=value", "session=abc123"])
+    pub cookies: Option<Vec<String>>,
 }
 
 #[derive(Debug, Deserialize, JsonSchema)]