From 4112234371f21cd07494a5b6f3689fcb110561fe Mon Sep 17 00:00:00 2001
From: jaydenfyi <213395523+jaydenfyi@users.noreply.github.com>
Date: Sun, 18 Jan 2026 10:18:06 +0800
Subject: [PATCH] fix(cli): Output screenshot as base64 string when no path
 provided (#83)

* fix(cli): print screenshot base64 when no path

* chore(docs): update docs and SKILL.md

* add test for screenshot with path arg

* more minimal readme + skill change
---
 README.md                         |  2 +-
 cli/src/commands.rs               | 14 +++++++++++++-
 cli/src/output.rs                 |  5 +++++
 docs/src/app/quick-start/page.tsx |  3 ++-
 4 files changed, 21 insertions(+), 3 deletions(-)

diff --git a/README.md b/README.md
index 8c742b0..5fe49b1 100644
--- a/README.md
+++ b/README.md
@@ -74,7 +74,7 @@ agent-browser scroll <dir> [px]       # Scroll (up/down/left/right)
 agent-browser scrollintoview <sel>    # Scroll element into view (alias: scrollinto)
 agent-browser drag <src> <tgt>        # Drag and drop
 agent-browser upload <sel> <files>    # Upload files
-agent-browser screenshot [path]       # Take screenshot (--full for full page)
+agent-browser screenshot [path]       # Take screenshot (--full for full page, base64 png to stdout if no path)
 agent-browser pdf <path>              # Save as PDF
 agent-browser snapshot                # Accessibility tree with refs (best for AI)
 agent-browser eval <js>               # Run JavaScript
diff --git a/cli/src/commands.rs b/cli/src/commands.rs
index 25c2e0d..e14713e 100644
--- a/cli/src/commands.rs
+++ b/cli/src/commands.rs
@@ -273,7 +273,11 @@ pub fn parse_command(args: &[String], flags: &Flags) -> Result<Value, ParseError
 
         // === Screenshot/PDF ===
         "screenshot" => {
-            Ok(json!({ "id": id, "action": "screenshot", "path": rest.get(0), "fullPage": flags.full }))
+            let mut cmd = json!({ "id": id, "action": "screenshot", "fullPage": flags.full });
+            if let Some(path) = rest.get(0) {
+                cmd["path"] = json!(path);
+            }
+            Ok(cmd)
         }
         "pdf" => {
             let path = rest.get(0).ok_or_else(|| ParseError::MissingArguments {
@@ -1286,6 +1290,14 @@ mod tests {
     fn test_screenshot() {
         let cmd = parse_command(&args("screenshot"), &default_flags()).unwrap();
         assert_eq!(cmd["action"], "screenshot");
+        assert!(cmd.get("path").is_none());
+    }
+
+    #[test]
+    fn test_screenshot_path() {
+        let cmd = parse_command(&args("screenshot out.png"), &default_flags()).unwrap();
+        assert_eq!(cmd["action"], "screenshot");
+        assert_eq!(cmd["path"], "out.png");
     }
 
     #[test]
diff --git a/cli/src/output.rs b/cli/src/output.rs
index 31b98d8..77b4df1 100644
--- a/cli/src/output.rs
+++ b/cli/src/output.rs
@@ -208,6 +208,11 @@ pub fn print_response(resp: &Response, json_mode: bool) {
             println!("\x1b[32m✓\x1b[0m Screenshot saved to {}", path);
             return;
         }
+        // Screenshot base64
+        if let Some(base64) = data.get("base64").and_then(|v| v.as_str()) {
+            println!("{}", base64);
+            return;
+        }
         // Default success
         println!("\x1b[32m✓\x1b[0m Done");
     }
diff --git a/docs/src/app/quick-start/page.tsx b/docs/src/app/quick-start/page.tsx
index 825b5a3..00545c6 100644
--- a/docs/src/app/quick-start/page.tsx
+++ b/docs/src/app/quick-start/page.tsx
@@ -12,7 +12,8 @@ agent-browser snapshot                    # Get accessibility tree with refs
 agent-browser click @e2                   # Click by ref from snapshot
 agent-browser fill @e3 "test@example.com" # Fill by ref
 agent-browser get text @e1                # Get text by ref
-agent-browser screenshot page.png
+agent-browser screenshot                  # Base64 png to stdout
+agent-browser screenshot page.png         # Save to file
 agent-browser close`} />
 
         <h2>Traditional selectors</h2>