zerochat 7.8.0__tar.gz → 7.10.0__tar.gz

{zerochat-7.8.0/zerochat.egg-info → zerochat-7.10.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: zerochat
-Version: 7.8.0
+Version: 7.10.0
 Summary: Private local web client for AI models, tools and MCP services
 Author: Alberto Alday
 License-Expression: MIT

{zerochat-7.8.0 → zerochat-7.10.0}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "zerochat"
-version = "7.8.0"
+version = "7.10.0"
 description = "Private local web client for AI models, tools and MCP services"
 readme = "README.md"
 requires-python = ">=3.10"

{zerochat-7.8.0 → zerochat-7.10.0/zerochat.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: zerochat
-Version: 7.8.0
+Version: 7.10.0
 Summary: Private local web client for AI models, tools and MCP services
 Author: Alberto Alday
 License-Expression: MIT

{zerochat-7.8.0 → zerochat-7.10.0}/zerochat.py

@@ -12,8 +12,10 @@ Proporciona:
 from __future__ import annotations
 
 import argparse
+import ast
 import atexit
 import datetime
+import fnmatch
 import hmac
 import importlib.metadata
 import json
@@ -27,6 +29,7 @@ import shlex
 import signal
 import subprocess
 import sys
+import tempfile
 import threading
 import time
 import traceback
@@ -37,7 +40,7 @@ import webbrowser
 from http.server import BaseHTTPRequestHandler, ThreadingHTTPServer
 from pathlib import Path
 
-SOURCE_BACKEND_VERSION = "7.8.0"
+SOURCE_BACKEND_VERSION = "7.10.0"
 
 def _read_source_version(filename: str) -> str | None:
     """Lee la versión de un archivo del repositorio cuando se ejecuta desde fuentes."""
@@ -388,34 +391,54 @@ def check_version():
 # Herramientas Locales Core
 # ==============================================================================
 
-def list_directory(path: str = ".") -> str:
-    """Lista los archivos y carpetas de un directorio, sin recorrer subdirectorios."""
+def list_directory(path: str = ".", recursive: bool = False) -> str:
+    """List files and directories in a local directory with safe bounded recursive traversal."""
     try:
         target = Path(path).expanduser().resolve()
         if not target.exists():
-            return json.dumps({"success": False, "error": f"La ruta '{path}' no existe."}, ensure_ascii=False)
+            return json.dumps({"success": False, "error": f"Path '{path}' does not exist."}, ensure_ascii=False)
         if not target.is_dir():
-            return json.dumps({"success": False, "error": f"La ruta '{path}' no es un directorio."}, ensure_ascii=False)
+            return json.dumps({"success": False, "error": f"Path '{path}' is not a directory."}, ensure_ascii=False)
 
+        max_depth = 3 if recursive else 1
         entries = []
-        for entry in os.scandir(target):
+
+        def _scan(current_path: Path, current_depth: int):
+            if current_depth > max_depth or len(entries) >= 1000:
+                return
             try:
-                stat = entry.stat(follow_symlinks=False)
-                is_dir = entry.is_dir(follow_symlinks=False)
-                entries.append({
-                    "name": entry.name,
-                    "path": str(Path(entry.path).resolve()),
-                    "type": "directory" if is_dir else "file",
-                    "size_bytes": None if is_dir else stat.st_size,
-                    "is_symlink": entry.is_symlink()
-                })
+                with os.scandir(current_path) as it:
+                    items = list(it)
+                    items.sort(key=lambda e: (not e.is_dir(follow_symlinks=False), e.name.lower()))
+                    for entry in items:
+                        if len(entries) >= 1000:
+                            break
+                        try:
+                            stat = entry.stat(follow_symlinks=False)
+                            is_dir = entry.is_dir(follow_symlinks=False)
+                            entry_data = {
+                                "name": entry.name,
+                                "path": str(Path(entry.path).resolve()),
+                                "type": "directory" if is_dir else "file",
+                                "size_bytes": None if is_dir else stat.st_size,
+                                "is_symlink": entry.is_symlink()
+                            }
+                            if recursive:
+                                entry_data["relative_path"] = str(Path(entry.path).resolve().relative_to(target))
+                            entries.append(entry_data)
+                            if recursive and is_dir and not entry.is_symlink():
+                                _scan(Path(entry.path), current_depth + 1)
+                        except (PermissionError, FileNotFoundError):
+                            continue
             except (PermissionError, FileNotFoundError):
-                continue
+                pass
+
+        _scan(target, 1)
 
-        entries.sort(key=lambda e: (e["type"] != "directory", e["name"].lower()))
         return json.dumps({
             "success": True,
             "path": str(target),
+            "recursive": bool(recursive),
             "total_items": len(entries),
             "entries": entries
         }, ensure_ascii=False, indent=2)
@@ -423,34 +446,65 @@ def list_directory(path: str = ".") -> str:
         return json.dumps({"success": False, "error": str(e)}, ensure_ascii=False)
 
 
-def read_file(path: str, start_line: int = 1, max_lines: int = 500, max_bytes: int = 100000) -> str:
-    """Lee el contenido de texto de un archivo local con rangos y límites seguros."""
+def read_file(path: str, start_line: int = 1, end_line: int = None, max_lines: int = 500, max_bytes: int = 100000) -> str:
+    """Read text content from a local file in streaming mode with safe ranges and bounds."""
     try:
         target = Path(path).expanduser().resolve()
         if not target.exists():
-            return json.dumps({"success": False, "error": f"El archivo '{path}' no existe."}, ensure_ascii=False)
+            return json.dumps({"success": False, "error": f"File '{path}' does not exist."}, ensure_ascii=False)
         if not target.is_file():
-            return json.dumps({"success": False, "error": f"La ruta '{path}' no es un archivo regular."}, ensure_ascii=False)
+            return json.dumps({"success": False, "error": f"Path '{path}' is not a regular file."}, ensure_ascii=False)
 
         file_size = target.stat().st_size
         safe_max_bytes = max(1024, min(int(max_bytes), 2000000))
         safe_start_line = max(1, int(start_line))
-        safe_max_lines = max(1, min(int(max_lines), 2000))
+
+        if end_line is not None:
+            safe_end_line = max(safe_start_line, int(end_line))
+            target_line_count = safe_end_line - safe_start_line + 1
+        else:
+            safe_end_line = None
+            target_line_count = max(1, min(int(max_lines), 2000))
+
+        selected_lines = []
+        current_bytes = 0
+        truncated_bytes = False
+        total_lines = 0
+        reached_end_bound = False
 
         with open(target, "r", encoding="utf-8", errors="replace") as f:
-            lines = f.readlines()
+            for line_no, line in enumerate(f, 1):
+                total_lines = line_no
+                if line_no < safe_start_line:
+                    continue
+                if safe_end_line is not None and line_no > safe_end_line:
+                    reached_end_bound = True
+                    if line_no > safe_start_line + 50000:
+                        break
+                    continue
+                if safe_end_line is None and len(selected_lines) >= target_line_count:
+                    reached_end_bound = True
+                    if line_no > safe_start_line + 50000:
+                        break
+                    continue
 
-        total_lines = len(lines)
-        start_idx = safe_start_line - 1
-        end_idx = min(start_idx + safe_max_lines, total_lines)
+                if not reached_end_bound:
+                    line_bytes = len(line.encode("utf-8"))
+                    if current_bytes + line_bytes > safe_max_bytes:
+                        remaining_budget = max(0, safe_max_bytes - current_bytes)
+                        if remaining_budget > 0:
+                            encoded = line.encode("utf-8")[:remaining_budget]
+                            selected_lines.append(encoded.decode("utf-8", errors="ignore"))
+                        truncated_bytes = True
+                        reached_end_bound = True
+                    else:
+                        selected_lines.append(line)
+                        current_bytes += line_bytes
 
-        selected_lines = lines[start_idx:end_idx] if start_idx < total_lines else []
         content = "".join(selected_lines)
+        last_returned_line = safe_start_line + len(selected_lines) - 1 if selected_lines else safe_start_line - 1
 
-        truncated_bytes = False
-        if len(content.encode("utf-8")) > safe_max_bytes:
-            content = content[:safe_max_bytes]
-            truncated_bytes = True
+        is_truncated = truncated_bytes or (safe_end_line is not None and safe_end_line < total_lines) or (safe_end_line is None and (safe_start_line + target_line_count - 1) < total_lines)
 
         return json.dumps({
             "success": True,
@@ -458,44 +512,103 @@ def read_file(path: str, start_line: int = 1, max_lines: int = 500, max_bytes: i
             "size_bytes": file_size,
             "total_lines": total_lines,
             "start_line": safe_start_line,
+            "end_line": last_returned_line,
             "lines_returned": len(selected_lines),
-            "truncated": truncated_bytes or end_idx < total_lines,
+            "truncated": is_truncated,
             "content": content
         }, ensure_ascii=False, indent=2)
     except Exception as e:
         return json.dumps({"success": False, "error": str(e)}, ensure_ascii=False)
 
 
-def edit_file(path: str, content: str, mode: str = "write", target_content: str = None) -> str:
-    """Crea, sobrescribe o edita un archivo de forma atómica."""
+def write_file(path: str, content: str) -> str:
+    """Create or overwrite a file completely and atomically."""
     try:
         target = Path(path).expanduser().resolve()
         target.parent.mkdir(parents=True, exist_ok=True)
+        temp_path = target.with_suffix(target.suffix + f".tmp_{os.getpid()}_{time.time_ns()}")
+        with open(temp_path, "w", encoding="utf-8") as f:
+            f.write(content)
+        temp_path.replace(target)
+        bytes_written = len(content.encode("utf-8"))
+        return json.dumps({
+            "success": True,
+            "path": str(target),
+            "bytes_written": bytes_written
+        }, ensure_ascii=False, indent=2)
+    except Exception as e:
+        return json.dumps({"success": False, "error": str(e)}, ensure_ascii=False)
+
+
+def edit_file(path: str, old_str: str = None, new_str: str = None, content: str = None, mode: str = "surgical", target_content: str = None) -> str:
+    """Modify an existing file surgically or atomically."""
+    try:
+        target = Path(path).expanduser().resolve()
+
+        # Preferred surgical mode (old_str -> new_str)
+        if old_str is not None and new_str is not None:
+            if not target.exists():
+                return json.dumps({"success": False, "error": f"File '{path}' does not exist. Use read_file to verify existing paths."}, ensure_ascii=False)
+            if not target.is_file():
+                return json.dumps({"success": False, "error": f"Path '{path}' is not a regular file."}, ensure_ascii=False)
 
+            with open(target, "r", encoding="utf-8", errors="replace") as f:
+                file_text = f.read()
+
+            occurrences = file_text.count(old_str)
+            if occurrences == 0:
+                return json.dumps({
+                    "success": False,
+                    "error": f"Target text was not found in '{path}'. Check exact file content with read_file."
+                }, ensure_ascii=False)
+            if occurrences > 1:
+                return json.dumps({
+                    "success": False,
+                    "error": f"Found {occurrences} matches for the snippet in '{path}'. Provide more surrounding context in old_str to ensure a unique match."
+                }, ensure_ascii=False)
+
+            updated_text = file_text.replace(old_str, new_str, 1)
+            temp_path = target.with_suffix(target.suffix + f".tmp_{os.getpid()}_{time.time_ns()}")
+            with open(temp_path, "w", encoding="utf-8") as f:
+                f.write(updated_text)
+            temp_path.replace(target)
+            return json.dumps({
+                "success": True,
+                "path": str(target),
+                "mode": "surgical",
+                "bytes_written": len(updated_text.encode("utf-8")),
+                "replacements": 1
+            }, ensure_ascii=False, indent=2)
+
+        # Legacy modes (write, append, replace_chunk)
+        if content is None:
+            return json.dumps({"success": False, "error": "Must provide old_str and new_str for surgical edit, or content for compatible modes."}, ensure_ascii=False)
+
+        target.parent.mkdir(parents=True, exist_ok=True)
         if mode == "append":
             with open(target, "a", encoding="utf-8") as f:
                 f.write(content)
             bytes_written = len(content.encode("utf-8"))
         elif mode == "replace_chunk":
             if not target.exists():
-                return json.dumps({"success": False, "error": f"El archivo '{path}' no existe para replace_chunk."}, ensure_ascii=False)
+                return json.dumps({"success": False, "error": f"File '{path}' does not exist for replace_chunk."}, ensure_ascii=False)
             if not target_content:
-                return json.dumps({"success": False, "error": "target_content es obligatorio en modo replace_chunk."}, ensure_ascii=False)
+                return json.dumps({"success": False, "error": "target_content is required in replace_chunk mode."}, ensure_ascii=False)
 
             with open(target, "r", encoding="utf-8", errors="replace") as f:
                 existing = f.read()
 
             if target_content not in existing:
-                return json.dumps({"success": False, "error": "target_content no fue encontrado en el archivo."}, ensure_ascii=False)
+                return json.dumps({"success": False, "error": "target_content was not found in the file."}, ensure_ascii=False)
 
             new_text = existing.replace(target_content, content, 1)
-            temp_path = target.with_suffix(target.suffix + f".tmp_{os.getpid()}")
+            temp_path = target.with_suffix(target.suffix + f".tmp_{os.getpid()}_{time.time_ns()}")
             with open(temp_path, "w", encoding="utf-8") as f:
                 f.write(new_text)
             temp_path.replace(target)
             bytes_written = len(new_text.encode("utf-8"))
         else:  # write
-            temp_path = target.with_suffix(target.suffix + f".tmp_{os.getpid()}")
+            temp_path = target.with_suffix(target.suffix + f".tmp_{os.getpid()}_{time.time_ns()}")
             with open(temp_path, "w", encoding="utf-8") as f:
                 f.write(content)
             temp_path.replace(target)
@@ -511,8 +624,135 @@ def edit_file(path: str, content: str, mode: str = "write", target_content: str
         return json.dumps({"success": False, "error": str(e)}, ensure_ascii=False)
 
 
+def truncate_terminal_output(text: str, max_chars: int = 8000, head_lines: int = 50, tail_lines: int = 30) -> tuple[str, bool]:
+    """Outputs >8,000 characters are truncated to first 50 lines + warning + last 30 lines."""
+    if len(text) <= max_chars:
+        return text, False
+    lines = text.splitlines(keepends=True)
+    if len(lines) <= head_lines + tail_lines:
+        half = max_chars // 2
+        return (
+            text[:half]
+            + f"\n\n[... Output truncated due to length ({len(text)} total characters) ...]\n\n"
+            + text[-half:],
+            True
+        )
+    omitted = len(lines) - head_lines - tail_lines
+    head = "".join(lines[:head_lines])
+    tail = "".join(lines[-tail_lines:])
+    warning = f"\n\n[... Output truncated: omitted {omitted} lines ({len(text)} total characters) ...]\n\n"
+    return head + warning + tail, True
+
+
+class PersistentBashSession:
+    """Maintains working directory (cwd) and environment variables across successive calls."""
+    def __init__(self):
+        self._dir = tempfile.mkdtemp(prefix="zerochat_bash_")
+        self._cwd_file = Path(self._dir) / "cwd"
+        self._env_file = Path(self._dir) / "env.sh"
+        self._cwd = str(Path.cwd().resolve())
+        self._lock = threading.Lock()
+        atexit.register(self.cleanup)
+
+    def cleanup(self):
+        try:
+            shutil.rmtree(self._dir, ignore_errors=True)
+        except Exception:
+            pass
+
+    def run(self, command: str, timeout_seconds: int = 30) -> str:
+        with self._lock:
+            safe_timeout = max(1, min(int(timeout_seconds), 300))
+            runner_script = Path(self._dir) / f"runner_{time.time_ns()}.sh"
+
+            script_content = (
+                "if [ -f " + shlex.quote(str(self._env_file)) + " ]; then\n"
+                "  . " + shlex.quote(str(self._env_file)) + " 2>/dev/null || true\n"
+                "fi\n"
+                "cd " + shlex.quote(self._cwd) + " 2>/dev/null || true\n"
+                "trap '__ret=$?; pwd > " + shlex.quote(str(self._cwd_file)) + "; export -p > " + shlex.quote(str(self._env_file)) + " 2>/dev/null || true; exit $__ret' EXIT\n"
+                + command + "\n"
+            )
+
+            try:
+                with open(runner_script, "w", encoding="utf-8") as f:
+                    f.write(script_content)
+                runner_script.chmod(0o700)
+
+                proc = subprocess.Popen(
+                    ["bash", str(runner_script)],
+                    cwd=self._cwd,
+                    stdout=subprocess.PIPE,
+                    stderr=subprocess.PIPE,
+                    text=True,
+                    encoding="utf-8",
+                    errors="replace",
+                    start_new_session=True if hasattr(os, "setsid") else False
+                )
+
+                try:
+                    stdout, stderr = proc.communicate(timeout=safe_timeout)
+                    returncode = proc.returncode
+                except subprocess.TimeoutExpired:
+                    try:
+                        if hasattr(os, "killpg") and hasattr(os, "getpgid"):
+                            os.killpg(os.getpgid(proc.pid), signal.SIGKILL)
+                        else:
+                            proc.kill()
+                    except Exception:
+                        try:
+                            proc.kill()
+                        except Exception:
+                            pass
+                    proc.communicate()
+                    return json.dumps({
+                        "success": False,
+                        "error": f"Command timed out after {timeout_seconds} seconds (terminated with SIGKILL).",
+                        "cwd": self._cwd
+                    }, ensure_ascii=False)
+
+                if self._cwd_file.is_file():
+                    try:
+                        saved_cwd = self._cwd_file.read_text(encoding="utf-8").strip()
+                        if saved_cwd and Path(saved_cwd).is_dir():
+                            self._cwd = saved_cwd
+                    except Exception:
+                        pass
+
+                truncated_out, was_out_trunc = truncate_terminal_output(stdout)
+                truncated_err, was_err_trunc = truncate_terminal_output(stderr)
+
+                return json.dumps({
+                    "success": returncode == 0,
+                    "returncode": returncode,
+                    "stdout": truncated_out,
+                    "stderr": truncated_err,
+                    "cwd": self._cwd,
+                    "truncated": was_out_trunc or was_err_trunc
+                }, ensure_ascii=False, indent=2)
+
+            except Exception as e:
+                return json.dumps({"success": False, "error": str(e), "cwd": self._cwd}, ensure_ascii=False)
+            finally:
+                try:
+                    if runner_script.is_file():
+                        runner_script.unlink()
+                except Exception:
+                    pass
+
+
+BASH_SESSION = PersistentBashSession()
+
+
+def bash(command: str, timeout_seconds: int = 30) -> str:
+    """Execute a command in an interactive persistent bash shell session."""
+    return BASH_SESSION.run(command, timeout_seconds=timeout_seconds)
+
+
 def execute_command(command: str, cwd: str = ".", timeout_seconds: int = 60) -> str:
-    """Ejecuta un comando en la shell del sistema y devuelve stdout y stderr."""
+    """Execute a command in the system shell and capture stdout and stderr."""
+    if cwd == "." or cwd == BASH_SESSION._cwd:
+        return BASH_SESSION.run(command, timeout_seconds=timeout_seconds)
     try:
         target_cwd = Path(cwd).expanduser().resolve()
         if not target_cwd.exists() or not target_cwd.is_dir():
@@ -528,15 +768,417 @@ def execute_command(command: str, cwd: str = ".", timeout_seconds: int = 60) ->
             encoding="utf-8",
             errors="replace"
         )
+        trunc_out, was_out_trunc = truncate_terminal_output(proc.stdout)
+        trunc_err, was_err_trunc = truncate_terminal_output(proc.stderr)
         return json.dumps({
             "success": proc.returncode == 0,
             "returncode": proc.returncode,
-            "stdout": proc.stdout,
-            "stderr": proc.stderr,
-            "cwd": str(target_cwd)
+            "stdout": trunc_out,
+            "stderr": trunc_err,
+            "cwd": str(target_cwd),
+            "truncated": was_out_trunc or was_err_trunc
         }, ensure_ascii=False, indent=2)
     except subprocess.TimeoutExpired:
-        return json.dumps({"success": False, "error": f"Comando excedió el tiempo límite de {timeout_seconds} segundos."}, ensure_ascii=False)
+        return json.dumps({"success": False, "error": f"Command timed out after {timeout_seconds} seconds."}, ensure_ascii=False)
+    except Exception as e:
+        return json.dumps({"success": False, "error": str(e)}, ensure_ascii=False)
+
+
+def search_files(query: str, path: str = ".", file_pattern: str = None, max_results: int = 100) -> str:
+    """Recursively search for plain text or regular expressions across project files."""
+    try:
+        target = Path(path).expanduser().resolve()
+        if not target.exists():
+            return json.dumps({"success": False, "error": f"Path '{path}' does not exist."}, ensure_ascii=False)
+        if not target.is_dir():
+            return json.dumps({"success": False, "error": f"Path '{path}' is not a directory."}, ensure_ascii=False)
+
+        try:
+            regex = re.compile(query, re.MULTILINE)
+        except re.error:
+            regex = re.compile(re.escape(query), re.MULTILINE)
+
+        ignored_dirs = {".git", ".venv", "node_modules", "__pycache__", ".pytest_cache", ".cache"}
+        matches = []
+        files_searched = 0
+        max_file_size = 2 * 1024 * 1024
+        truncated = False
+
+        for root, dirs, files in os.walk(target):
+            dirs[:] = [d for d in dirs if d not in ignored_dirs and not d.startswith(".")]
+
+            for fname in sorted(files):
+                if file_pattern and not fnmatch.fnmatch(fname, file_pattern):
+                    continue
+
+                fpath = Path(root) / fname
+                try:
+                    stat = fpath.stat()
+                    if stat.st_size > max_file_size:
+                        continue
+                except OSError:
+                    continue
+
+                files_searched += 1
+                try:
+                    with open(fpath, "r", encoding="utf-8", errors="ignore") as f:
+                        for line_idx, line in enumerate(f, 1):
+                            if regex.search(line):
+                                try:
+                                    rel_path = str(fpath.relative_to(target))
+                                except ValueError:
+                                    rel_path = str(fpath)
+                                matches.append({
+                                    "file": str(fpath),
+                                    "relative_path": rel_path,
+                                    "line_number": line_idx,
+                                    "content": line.rstrip("\r\n")[:300]
+                                })
+                                if len(matches) >= max_results:
+                                    truncated = True
+                                    break
+                except (PermissionError, OSError):
+                    continue
+
+                if truncated:
+                    break
+            if truncated:
+                break
+
+        return json.dumps({
+            "success": True,
+            "path": str(target),
+            "query": query,
+            "file_pattern": file_pattern,
+            "files_searched": files_searched,
+            "total_matches": len(matches),
+            "truncated": truncated,
+            "matches": matches
+        }, ensure_ascii=False, indent=2)
+    except Exception as e:
+        return json.dumps({"success": False, "error": str(e)}, ensure_ascii=False)
+
+
+def get_diagnostics(path: str | None = None) -> str:
+    """Retrieve syntax diagnostics and code errors for a file or the entire workspace."""
+    try:
+        if path is not None and str(path).strip():
+            target = Path(str(path).strip()).expanduser().resolve()
+        else:
+            target = Path.cwd().resolve()
+
+        if not target.exists():
+            return json.dumps({
+                "success": False,
+                "error": f"Path '{path}' does not exist."
+            }, ensure_ascii=False)
+
+        diagnostics: list[dict] = []
+        files_checked = 0
+
+        def check_single_file(fpath: Path) -> list[dict]:
+            ext = fpath.suffix.lower()
+            file_diags: list[dict] = []
+            try:
+                rel_display = str(fpath.relative_to(Path.cwd()))
+            except ValueError:
+                rel_display = str(fpath)
+
+            if ext == ".py":
+                try:
+                    content = fpath.read_text(encoding="utf-8", errors="replace")
+                    compile(content, str(fpath), "exec")
+                except (SyntaxError, IndentationError) as e:
+                    file_diags.append({
+                        "file": rel_display,
+                        "line": e.lineno or 1,
+                        "column": e.offset or 1,
+                        "severity": "error",
+                        "message": f"SyntaxError: {e.msg}"
+                    })
+            elif ext in (".js", ".mjs", ".cjs"):
+                node_bin = shutil.which("node")
+                if node_bin:
+                    try:
+                        res = subprocess.run([node_bin, "--check", str(fpath)], capture_output=True, text=True, timeout=10)
+                        if res.returncode != 0:
+                            stderr = res.stderr or ""
+                            line_match = re.search(r':(\d+)\n([^\n]+)\n(\s*)\^', stderr)
+                            line = int(line_match.group(1)) if line_match else 1
+                            col = len(line_match.group(3)) + 1 if line_match else 1
+                            msg_match = re.search(r'(SyntaxError:[^\n]+)', stderr)
+                            msg = msg_match.group(1) if msg_match else (stderr.strip().splitlines()[-1] if stderr.strip() else "Syntax error")
+                            file_diags.append({
+                                "file": rel_display,
+                                "line": line,
+                                "column": col,
+                                "severity": "error",
+                                "message": msg
+                            })
+                    except Exception:
+                        pass
+            elif ext == ".json":
+                try:
+                    content = fpath.read_text(encoding="utf-8", errors="replace")
+                    json.loads(content)
+                except json.JSONDecodeError as e:
+                    file_diags.append({
+                        "file": rel_display,
+                        "line": e.lineno,
+                        "column": e.colno,
+                        "severity": "error",
+                        "message": f"JSONDecodeError: {e.msg}"
+                    })
+            return file_diags
+
+        if target.is_file():
+            files_checked = 1
+            diagnostics.extend(check_single_file(target))
+        else:
+            ignore_dirs = {".git", "node_modules", ".venv", "venv", "dist", "build", "__pycache__", ".gemini", ".cache"}
+            max_scan = 200
+            for root, dirs, files in os.walk(target):
+                dirs[:] = [d for d in dirs if d not in ignore_dirs and not d.startswith(".")]
+                for fname in sorted(files):
+                    fp = Path(root) / fname
+                    if fp.suffix.lower() in (".py", ".js", ".mjs", ".cjs", ".json"):
+                        files_checked += 1
+                        diagnostics.extend(check_single_file(fp))
+                        if files_checked >= max_scan:
+                            break
+                if files_checked >= max_scan:
+                    break
+
+        error_count = sum(1 for d in diagnostics if d.get("severity") == "error")
+        warning_count = sum(1 for d in diagnostics if d.get("severity") == "warning")
+        msg = f"Found {len(diagnostics)} issue(s)." if diagnostics else "No diagnostic issues found."
+
+        return json.dumps({
+            "success": True,
+            "path": str(target),
+            "files_checked": files_checked,
+            "error_count": error_count,
+            "warning_count": warning_count,
+            "diagnostics": diagnostics,
+            "message": msg
+        }, ensure_ascii=False, indent=2)
+    except Exception as e:
+        return json.dumps({"success": False, "error": str(e)}, ensure_ascii=False)
+
+
+class PersistentBrowserSession:
+    """Manages a persistent headless browser session (Playwright) for browser_action."""
+
+    def __init__(self):
+        self._lock = threading.Lock()
+        self._process: subprocess.Popen | None = None
+        atexit.register(self.close)
+
+    def _ensure_running(self) -> subprocess.Popen:
+        if self._process is not None and self._process.poll() is None:
+            return self._process
+
+        node = shutil.which("node")
+        if not node:
+            raise RuntimeError("Node.js is not installed or not found in system PATH.")
+
+        runner_js = """
+const readline = require('readline');
+let playwright;
+try {
+  playwright = require('playwright');
+} catch (e1) {
+  try {
+    const path = require('path');
+    const home = process.env.HOME || process.env.USERPROFILE || '';
+    playwright = require(path.join(home, 'zerochat', 'services', 'playwright', 'node_modules', 'playwright'));
+  } catch (e2) {
+    console.log(JSON.stringify({
+      success: false,
+      error: "Playwright is not available. Install it with 'npm install playwright' or enable the Playwright MCP service."
+    }));
+    process.exit(1);
+  }
+}
+
+(async () => {
+  let browser, context, page;
+  try {
+    const candidates = [
+      {},                      // 1. Playwright bundled Chromium
+      { channel: 'chrome' },   // 2. System Google Chrome
+      { channel: 'msedge' },   // 3. System Microsoft Edge (Windows 10/11)
+      { channel: 'chromium' }  // 4. System Chromium (/usr/bin/chromium)
+    ];
+    let lastErr = null;
+    for (const cand of candidates) {
+      try {
+        browser = await playwright.chromium.launch({ ...cand, headless: true });
+        if (browser) break;
+      } catch (e) {
+        lastErr = e;
+      }
+    }
+    if (!browser) {
+      throw lastErr || new Error('No Chromium-based browser found (Playwright, Chrome, Edge, or Chromium).');
+    }
+    context = await browser.newContext({ viewport: { width: 1280, height: 800 } });
+    page = await context.newPage();
+  } catch (err) {
+    console.log(JSON.stringify({ success: false, error: 'Failed to start browser: ' + err.message }));
+    process.exit(1);
+  }
+
+  console.log(JSON.stringify({ ready: true }));
+
+  const rl = readline.createInterface({ input: process.stdin, output: process.stdout, terminal: false });
+  for await (const line of rl) {
+    if (!line.trim()) continue;
+    let req;
+    try {
+      req = JSON.parse(line);
+    } catch (e) {
+      console.log(JSON.stringify({ success: false, error: 'Invalid command JSON' }));
+      continue;
+    }
+
+    try {
+      const act = req.action;
+      if (act === 'navigate') {
+        if (!req.url) throw new Error("Parameter 'url' is required for action 'navigate'");
+        await page.goto(req.url, { waitUntil: 'domcontentloaded', timeout: 30000 });
+        console.log(JSON.stringify({
+          success: true,
+          action: 'navigate',
+          url: page.url(),
+          title: await page.title()
+        }));
+      } else if (act === 'screenshot') {
+        if (req.url && req.url !== page.url()) {
+          await page.goto(req.url, { waitUntil: 'domcontentloaded', timeout: 30000 });
+        }
+        const buf = await page.screenshot({ fullPage: false });
+        console.log(JSON.stringify({
+          success: true,
+          action: 'screenshot',
+          image_base64: buf.toString('base64'),
+          mime_type: 'image/png',
+          url: page.url(),
+          title: await page.title()
+        }));
+      } else if (act === 'click') {
+        if (!req.selector) throw new Error("Parameter 'selector' is required for action 'click'");
+        await page.click(req.selector, { timeout: 15000 });
+        console.log(JSON.stringify({
+          success: true,
+          action: 'click',
+          selector: req.selector,
+          url: page.url()
+        }));
+      } else if (act === 'fill') {
+        if (!req.selector) throw new Error("Parameter 'selector' is required for action 'fill'");
+        await page.fill(req.selector, req.value || '', { timeout: 15000 });
+        console.log(JSON.stringify({
+          success: true,
+          action: 'fill',
+          selector: req.selector,
+          value: req.value || '',
+          url: page.url()
+        }));
+      } else if (act === 'close') {
+        await browser.close();
+        console.log(JSON.stringify({ success: true, action: 'close' }));
+        process.exit(0);
+      } else {
+        console.log(JSON.stringify({ success: false, error: 'Unsupported action: ' + act }));
+      }
+    } catch (err) {
+      console.log(JSON.stringify({ success: false, error: err.message, action: req.action }));
+    }
+  }
+})();
+"""
+        proc = subprocess.Popen(
+            [node, "-e", runner_js],
+            stdin=subprocess.PIPE,
+            stdout=subprocess.PIPE,
+            stderr=subprocess.PIPE,
+            text=True,
+            encoding="utf-8",
+            bufsize=1
+        )
+        self._process = proc
+        first_line = proc.stdout.readline()
+        if not first_line:
+            err = proc.stderr.read()
+            self._process = None
+            raise RuntimeError(f"Failed to initialize headless browser: {err or 'process terminated unexpectedly'}")
+        data = json.loads(first_line)
+        if not data.get("ready"):
+            self._process = None
+            raise RuntimeError(data.get("error", "Unknown error starting browser"))
+
+        return proc
+
+    def execute(self, command: dict) -> dict:
+        with self._lock:
+            try:
+                proc = self._ensure_running()
+            except Exception as e:
+                return {"success": False, "error": str(e)}
+
+            req_json = json.dumps(command, ensure_ascii=False) + "\n"
+            try:
+                proc.stdin.write(req_json)
+                proc.stdin.flush()
+                resp_line = proc.stdout.readline()
+                if not resp_line:
+                    self.close()
+                    return {"success": False, "error": "Browser process closed unexpectedly."}
+                return json.loads(resp_line)
+            except Exception as e:
+                self.close()
+                return {"success": False, "error": f"Error executing browser action: {e}"}
+
+    def close(self):
+        with self._lock:
+            if self._process is not None:
+                try:
+                    if self._process.poll() is None:
+                        try:
+                            self._process.stdin.write(json.dumps({"action": "close"}) + "\n")
+                            self._process.stdin.flush()
+                            self._process.wait(timeout=2)
+                        except Exception:
+                            self._process.kill()
+                except Exception:
+                    pass
+                self._process = None
+
+
+_BROWSER_SESSION = PersistentBrowserSession()
+
+
+def browser_action(action: str, url: str | None = None, selector: str | None = None, value: str | None = None) -> str:
+    """Control a headless browser for UI testing and visual inspection."""
+    try:
+        act = (action or "").strip().lower()
+        if act not in ("navigate", "screenshot", "click", "fill"):
+            return json.dumps({
+                "success": False,
+                "error": f"Invalid browser action: '{action}'. Valid actions: navigate, screenshot, click, fill."
+            }, ensure_ascii=False)
+
+        cmd = {"action": act}
+        if url:
+            cmd["url"] = str(url).strip()
+        if selector:
+            cmd["selector"] = str(selector).strip()
+        if value is not None:
+            cmd["value"] = str(value)
+
+        result = _BROWSER_SESSION.execute(cmd)
+        return json.dumps(result, ensure_ascii=False, indent=2)
     except Exception as e:
         return json.dumps({"success": False, "error": str(e)}, ensure_ascii=False)
 
@@ -544,62 +1186,136 @@ def execute_command(command: str, cwd: str = ".", timeout_seconds: int = 60) ->
 LOCAL_TOOLS_DEFINITIONS = [
     {
         "name": "list_directory",
-        "description": "Lista archivos y carpetas de un directorio local, sin recorrer subdirectorios.",
+        "description": "List files and directories in a local directory with safe bounded recursive traversal.",
         "inputSchema": {
             "type": "object",
             "properties": {
-                "path": {"type": "string", "description": "Ruta relativa o absoluta (por defecto '.')"}
+                "path": {"type": "string", "description": "Relative or absolute directory path (defaults to '.')"},
+                "recursive": {"type": "boolean", "description": "If true, traverses subdirectories up to depth 3", "default": False}
             }
         }
     },
     {
         "name": "read_file",
-        "description": "Lee el contenido de texto de un archivo local.",
+        "description": "Read text content from a local file with safe ranges and bounds.",
         "inputSchema": {
             "type": "object",
             "properties": {
-                "path": {"type": "string", "description": "Ruta del archivo"},
-                "start_line": {"type": "integer", "description": "Línea inicial (1-indexed)", "default": 1},
-                "max_lines": {"type": "integer", "description": "Número máximo de líneas a leer", "default": 500},
-                "max_bytes": {"type": "integer", "description": "Límite máximo de bytes", "default": 100000}
+                "path": {"type": "string", "description": "Path to the file to read"},
+                "start_line": {"type": "integer", "description": "Starting line number (1-indexed)", "default": 1},
+                "end_line": {"type": "integer", "description": "Ending line number (inclusive)"},
+                "max_lines": {"type": "integer", "description": "Maximum number of lines to read when end_line is omitted", "default": 500},
+                "max_bytes": {"type": "integer", "description": "Maximum byte budget", "default": 100000}
             },
             "required": ["path"]
         }
     },
     {
-        "name": "edit_file",
-        "description": "Crea, sobrescribe o modifica un archivo local de forma atómica.",
+        "name": "write_file",
+        "description": "Create or overwrite a file completely and atomically.",
         "inputSchema": {
             "type": "object",
             "properties": {
-                "path": {"type": "string", "description": "Ruta del archivo a editar"},
-                "content": {"type": "string", "description": "Contenido a escribir o reemplazar"},
-                "mode": {"type": "string", "enum": ["write", "append", "replace_chunk"], "default": "write"},
-                "target_content": {"type": "string", "description": "Texto exacto a reemplazar cuando mode='replace_chunk'"}
+                "path": {"type": "string", "description": "Path to the file to write"},
+                "content": {"type": "string", "description": "Full content of the file"}
             },
             "required": ["path", "content"]
         }
     },
+    {
+        "name": "edit_file",
+        "description": "Edit an existing file surgically by replacing old_str with new_str (must match exactly once).",
+        "inputSchema": {
+            "type": "object",
+            "properties": {
+                "path": {"type": "string", "description": "Path to the file to edit"},
+                "old_str": {"type": "string", "description": "Exact snippet to replace (must appear exactly once in the file)"},
+                "new_str": {"type": "string", "description": "Replacement snippet"},
+                "content": {"type": "string", "description": "Content to write in legacy/compatible mode"},
+                "mode": {"type": "string", "enum": ["surgical", "write", "append", "replace_chunk"], "default": "surgical"},
+                "target_content": {"type": "string", "description": "Exact target content in replace_chunk mode"}
+            },
+            "required": ["path"]
+        }
+    },
+    {
+        "name": "bash",
+        "description": "Execute a command in an interactive persistent shell session (preserves cwd and exported environment variables across calls).",
+        "inputSchema": {
+            "type": "object",
+            "properties": {
+                "command": {"type": "string", "description": "Shell command to execute (e.g. npm test, git status)"},
+                "timeout_seconds": {"type": "integer", "description": "Timeout in seconds (defaults to 30)", "default": 30}
+            },
+            "required": ["command"]
+        }
+    },
+    {
+        "name": "search_files",
+        "description": "Recursively search for plain text or regular expressions across project files.",
+        "inputSchema": {
+            "type": "object",
+            "properties": {
+                "query": {"type": "string", "description": "Plain text string or regular expression to search for"},
+                "path": {"type": "string", "description": "Base search directory (defaults to '.')", "default": "."},
+                "file_pattern": {"type": "string", "description": "Optional glob filter (e.g. *.js, *.py)"}
+            },
+            "required": ["query"]
+        }
+    },
     {
         "name": "execute_command",
-        "description": "Ejecuta un comando en la shell del sistema y captura la salida.",
+        "description": "Execute a shell command and capture stdout/stderr.",
         "inputSchema": {
             "type": "object",
             "properties": {
-                "command": {"type": "string", "description": "Comando a ejecutar"},
-                "cwd": {"type": "string", "description": "Directorio de trabajo (por defecto '.')"},
-                "timeout_seconds": {"type": "integer", "description": "Tiempo límite en segundos", "default": 60}
+                "command": {"type": "string", "description": "Command to execute"},
+                "cwd": {"type": "string", "description": "Working directory (defaults to '.')"},
+                "timeout_seconds": {"type": "integer", "description": "Timeout in seconds (defaults to 60)", "default": 60}
             },
             "required": ["command"]
         }
+    },
+    {
+        "name": "get_diagnostics",
+        "description": "Retrieve syntax diagnostics and code errors for a file or the entire workspace.",
+        "inputSchema": {
+            "type": "object",
+            "properties": {
+                "path": {"type": "string", "description": "Path to the file or directory to inspect (defaults to '.')"}
+            }
+        }
+    },
+    {
+        "name": "browser_action",
+        "description": "Control a headless browser (Playwright) for web navigation, UI testing, and visual inspection via screenshots.",
+        "inputSchema": {
+            "type": "object",
+            "properties": {
+                "action": {
+                    "type": "string",
+                    "enum": ["navigate", "screenshot", "click", "fill"],
+                    "description": "Action to perform in the browser"
+                },
+                "url": {"type": "string", "description": "Target URL for navigate or screenshot"},
+                "selector": {"type": "string", "description": "CSS selector for click or fill actions"},
+                "value": {"type": "string", "description": "Text value to type for fill action"}
+            },
+            "required": ["action"]
+        }
     }
 ]
 
 LOCAL_TOOL_HANDLERS = {
     "list_directory": list_directory,
     "read_file": read_file,
+    "write_file": write_file,
     "edit_file": edit_file,
-    "execute_command": execute_command
+    "bash": bash,
+    "search_files": search_files,
+    "execute_command": execute_command,
+    "get_diagnostics": get_diagnostics,
+    "browser_action": browser_action
 }
 
 # ==============================================================================
@@ -616,18 +1332,51 @@ MAX_COMMAND_LENGTH = 16384
 def validate_local_tool_arguments(tool_name: str, arguments: dict) -> str | None:
     """Valida tipos y límites de las herramientas locales antes de ejecutarlas."""
     schemas = {
-        "list_directory": {"path": (str, MAX_PATH_LENGTH)},
+        "list_directory": {
+            "path": (str, MAX_PATH_LENGTH),
+            "recursive": (bool, None)
+        },
         "read_file": {
-            "path": (str, MAX_PATH_LENGTH), "start_line": (int, None),
-            "max_lines": (int, None), "max_bytes": (int, None)
+            "path": (str, MAX_PATH_LENGTH),
+            "start_line": (int, None),
+            "end_line": (int, None),
+            "max_lines": (int, None),
+            "max_bytes": (int, None)
+        },
+        "write_file": {
+            "path": (str, MAX_PATH_LENGTH),
+            "content": (str, MAX_HTTP_BODY_BYTES)
         },
         "edit_file": {
-            "path": (str, MAX_PATH_LENGTH), "content": (str, MAX_HTTP_BODY_BYTES),
-            "mode": (str, 32), "target_content": (str, MAX_HTTP_BODY_BYTES)
+            "path": (str, MAX_PATH_LENGTH),
+            "old_str": (str, MAX_HTTP_BODY_BYTES),
+            "new_str": (str, MAX_HTTP_BODY_BYTES),
+            "content": (str, MAX_HTTP_BODY_BYTES),
+            "mode": (str, 32),
+            "target_content": (str, MAX_HTTP_BODY_BYTES)
+        },
+        "bash": {
+            "command": (str, MAX_COMMAND_LENGTH),
+            "timeout_seconds": (int, None)
+        },
+        "search_files": {
+            "query": (str, 4096),
+            "path": (str, MAX_PATH_LENGTH),
+            "file_pattern": (str, 256)
         },
         "execute_command": {
-            "command": (str, MAX_COMMAND_LENGTH), "cwd": (str, MAX_PATH_LENGTH),
+            "command": (str, MAX_COMMAND_LENGTH),
+            "cwd": (str, MAX_PATH_LENGTH),
             "timeout_seconds": (int, None)
+        },
+        "get_diagnostics": {
+            "path": (str, MAX_PATH_LENGTH)
+        },
+        "browser_action": {
+            "action": (str, 32),
+            "url": (str, 4096),
+            "selector": (str, 1024),
+            "value": (str, MAX_HTTP_BODY_BYTES)
         }
     }
     schema = schemas.get(tool_name)
@@ -639,19 +1388,38 @@ def validate_local_tool_arguments(tool_name: str, arguments: dict) -> str | None
         if expected is None:
             return f"Argumento no permitido: {name}"
         expected_type, max_length = expected
-        if isinstance(value, bool) or not isinstance(value, expected_type):
-            return f"Tipo inválido para '{name}'"
+        if expected_type is bool:
+            if not isinstance(value, bool):
+                return f"Tipo inválido para '{name}'"
+        else:
+            if isinstance(value, bool) or not isinstance(value, expected_type):
+                return f"Tipo inválido para '{name}'"
         if max_length is not None and len(value) > max_length:
             return f"'{name}' excede el tamaño máximo permitido"
 
     required = {
         "read_file": ("path",),
-        "edit_file": ("path", "content"),
-        "execute_command": ("command",)
+        "write_file": ("path", "content"),
+        "edit_file": ("path",),
+        "bash": ("command",),
+        "search_files": ("query",),
+        "execute_command": ("command",),
+        "browser_action": ("action",)
     }
     for name in required.get(tool_name, ()):
         if name not in arguments:
             return f"Falta el argumento obligatorio: {name}"
+
+    if tool_name == "edit_file":
+        has_surgical = "old_str" in arguments and "new_str" in arguments
+        has_legacy = "content" in arguments
+        if not has_surgical and not has_legacy:
+            return "Faltan argumentos obligatorios: especifica 'old_str' y 'new_str' o 'content'"
+
+    if tool_name == "browser_action":
+        if arguments.get("action") not in ("navigate", "screenshot", "click", "fill"):
+            return f"Acción de navegador no permitida: {arguments.get('action')}"
+
     return None
 
 def sanitize_log_path(raw_path: str) -> str:
@@ -1590,6 +2358,12 @@ class ZeroChatServerHandler(BaseHTTPRequestHandler):
             return
         if content_len > MAX_HTTP_BODY_BYTES:
             self._log_req("POST", safe_path)
+            if content_len <= MAX_HTTP_BODY_BYTES * 2:
+                try:
+                    self.rfile.read(content_len)
+                except Exception:
+                    pass
+            self.close_connection = True
             self._send_json_response(413, {"error": "Request body too large"})
             self._log_res(413, safe_path, (time.monotonic() - t0) * 1000, "Cuerpo HTTP excede el límite")
             return