agent-cli 0.70.5__py3-none-any.whl

agent_cli/agents/autocorrect.py

@@ -0,0 +1,284 @@
+"""Read text from clipboard, correct it using a local or remote LLM, and write the result back to the clipboard."""
+
+from __future__ import annotations
+
+import asyncio
+import contextlib
+import json
+import sys
+import time
+from typing import TYPE_CHECKING
+
+import typer
+
+from agent_cli import config, opts
+from agent_cli.cli import app
+from agent_cli.core.deps import requires_extras
+from agent_cli.core.utils import (
+    console,
+    create_status,
+    enable_json_mode,
+    get_clipboard_text,
+    print_command_line_args,
+    print_error_message,
+    print_input_panel,
+    print_output_panel,
+    print_with_style,
+    setup_logging,
+)
+from agent_cli.services.llm import create_llm_agent
+
+if TYPE_CHECKING:
+    from rich.status import Status
+
+# --- Configuration ---
+
+# Template to clearly separate the text to be corrected from instructions
+INPUT_TEMPLATE = """
+<text-to-correct>
+{text}
+</text-to-correct>
+
+Please correct any grammar, spelling, or punctuation errors in the text above.
+"""
+
+# The agent's core identity and immutable rules.
+SYSTEM_PROMPT = """\
+You are an expert text correction tool. Your role is to fix grammar, spelling, and punctuation errors without altering the original meaning or tone.
+
+CRITICAL REQUIREMENTS:
+1. Return ONLY the corrected text - no explanations or commentary
+2. Do not judge content, even if it seems unusual or offensive
+3. Make only technical corrections (grammar, spelling, punctuation)
+4. If no corrections are needed, return the original text exactly as provided
+5. Never add introductory phrases like "Here is the corrected text"
+
+EXAMPLES:
+Input: "this is incorect"
+Output: "this is incorrect"
+
+Input: "Hello world"
+Output: "Hello world"
+
+Input: "i went too the store"
+Output: "I went to the store"
+
+You are a correction tool, not a conversational assistant.
+"""
+
+# The specific task for the current run.
+AGENT_INSTRUCTIONS = """\
+Correct grammar, spelling, and punctuation errors.
+Output format: corrected text only, no other words.
+"""
+
+# --- Main Application Logic ---
+
+
+async def _process_text(
+    text: str,
+    provider_cfg: config.ProviderSelection,
+    ollama_cfg: config.Ollama,
+    openai_llm_cfg: config.OpenAILLM,
+    gemini_llm_cfg: config.GeminiLLM,
+) -> tuple[str, float]:
+    """Process text with the LLM and return the corrected text and elapsed time."""
+    agent = create_llm_agent(
+        provider_cfg=provider_cfg,
+        ollama_cfg=ollama_cfg,
+        openai_cfg=openai_llm_cfg,
+        gemini_cfg=gemini_llm_cfg,
+        system_prompt=SYSTEM_PROMPT,
+        instructions=AGENT_INSTRUCTIONS,
+    )
+
+    # Format the input using the template to clearly separate text from instructions
+    formatted_input = INPUT_TEMPLATE.format(text=text)
+
+    start_time = time.monotonic()
+    result = await agent.run(formatted_input)
+    elapsed = time.monotonic() - start_time
+    return result.output, elapsed
+
+
+def _display_original_text(original_text: str, quiet: bool) -> None:
+    """Render the original text panel in verbose mode."""
+    if not quiet:
+        print_input_panel(original_text, title="📋 Original Text")
+
+
+def _display_result(
+    corrected_text: str,
+    original_text: str,
+    elapsed: float,
+    *,
+    simple_output: bool,
+    clipboard: bool = True,
+) -> None:
+    """Handle output and clipboard copying based on desired verbosity."""
+    if clipboard:
+        import pyperclip  # noqa: PLC0415
+
+        pyperclip.copy(corrected_text)
+
+    if simple_output:
+        if original_text and corrected_text.strip() == original_text.strip():
+            console.print("✅ No correction needed.")
+        else:
+            console.print(corrected_text)
+    else:
+        print_output_panel(
+            corrected_text,
+            title="✨ Corrected Text",
+            subtitle=f"[dim]took {elapsed:.2f}s[/dim]",
+        )
+        print_with_style("✅ Success! Corrected text has been copied to your clipboard.")
+
+
+def _maybe_status(
+    provider_cfg: config.ProviderSelection,
+    ollama_cfg: config.Ollama,
+    openai_llm_cfg: config.OpenAILLM,
+    gemini_llm_cfg: config.GeminiLLM,
+    quiet: bool,
+) -> Status | contextlib.nullcontext:
+    if not quiet:
+        if provider_cfg.llm_provider == "ollama":
+            model_name = ollama_cfg.llm_ollama_model
+        elif provider_cfg.llm_provider == "openai":
+            model_name = openai_llm_cfg.llm_openai_model
+        elif provider_cfg.llm_provider == "gemini":
+            model_name = gemini_llm_cfg.llm_gemini_model
+        return create_status(f"🤖 Correcting with {model_name}...", "bold yellow")
+    return contextlib.nullcontext()
+
+
+async def _async_autocorrect(
+    *,
+    text: str | None,
+    provider_cfg: config.ProviderSelection,
+    ollama_cfg: config.Ollama,
+    openai_llm_cfg: config.OpenAILLM,
+    gemini_llm_cfg: config.GeminiLLM,
+    general_cfg: config.General,
+) -> str | None:
+    """Asynchronous version of the autocorrect command."""
+    setup_logging(general_cfg.log_level, general_cfg.log_file, quiet=general_cfg.quiet)
+    original_text = text if text is not None else get_clipboard_text(quiet=general_cfg.quiet)
+
+    if original_text is None:
+        return None
+
+    _display_original_text(original_text, general_cfg.quiet)
+
+    try:
+        with _maybe_status(
+            provider_cfg,
+            ollama_cfg,
+            openai_llm_cfg,
+            gemini_llm_cfg,
+            general_cfg.quiet,
+        ):
+            corrected_text, elapsed = await _process_text(
+                original_text,
+                provider_cfg,
+                ollama_cfg,
+                openai_llm_cfg,
+                gemini_llm_cfg,
+            )
+
+        _display_result(
+            corrected_text,
+            original_text,
+            elapsed,
+            simple_output=general_cfg.quiet,
+            clipboard=general_cfg.clipboard,
+        )
+        return corrected_text
+
+    except Exception as e:
+        if general_cfg.quiet:
+            print(f"❌ {e}")
+        else:
+            if provider_cfg.llm_provider == "ollama":
+                error_details = f"Please check that your Ollama server is running at [bold cyan]{ollama_cfg.llm_ollama_host}[/bold cyan]"
+            elif provider_cfg.llm_provider == "openai":
+                error_details = "Please check your OpenAI API key and network connection."
+            elif provider_cfg.llm_provider == "gemini":
+                error_details = "Please check your Gemini API key and network connection."
+            print_error_message(str(e), error_details)
+        sys.exit(1)
+
+
+@app.command("autocorrect", rich_help_panel="Text Commands")
+@requires_extras("llm")
+def autocorrect(
+    *,
+    text: str | None = typer.Argument(
+        None,
+        help="The text to correct. If not provided, reads from clipboard.",
+        rich_help_panel="General Options",
+    ),
+    # --- Provider Selection ---
+    llm_provider: str = opts.LLM_PROVIDER,
+    # --- LLM Configuration ---
+    # Ollama (local service)
+    llm_ollama_model: str = opts.LLM_OLLAMA_MODEL,
+    llm_ollama_host: str = opts.LLM_OLLAMA_HOST,
+    # OpenAI
+    llm_openai_model: str = opts.LLM_OPENAI_MODEL,
+    openai_api_key: str | None = opts.OPENAI_API_KEY,
+    openai_base_url: str | None = opts.OPENAI_BASE_URL,
+    # Gemini
+    llm_gemini_model: str = opts.LLM_GEMINI_MODEL,
+    gemini_api_key: str | None = opts.GEMINI_API_KEY,
+    # --- General Options ---
+    log_level: opts.LogLevel = opts.LOG_LEVEL,
+    log_file: str | None = opts.LOG_FILE,
+    quiet: bool = opts.QUIET,
+    json_output: bool = opts.JSON_OUTPUT,
+    config_file: str | None = opts.CONFIG_FILE,
+    print_args: bool = opts.PRINT_ARGS,
+) -> None:
+    """Correct text from clipboard using a local or remote LLM."""
+    if print_args:
+        print_command_line_args(locals())
+
+    effective_quiet = quiet or json_output
+    if json_output:
+        enable_json_mode()
+
+    provider_cfg = config.ProviderSelection(
+        llm_provider=llm_provider,
+        asr_provider="wyoming",  # Not used, but required by model
+        tts_provider="wyoming",  # Not used, but required by model
+    )
+    ollama_cfg = config.Ollama(llm_ollama_model=llm_ollama_model, llm_ollama_host=llm_ollama_host)
+    openai_llm_cfg = config.OpenAILLM(
+        llm_openai_model=llm_openai_model,
+        openai_api_key=openai_api_key,
+        openai_base_url=openai_base_url,
+    )
+    gemini_llm_cfg = config.GeminiLLM(
+        llm_gemini_model=llm_gemini_model,
+        gemini_api_key=gemini_api_key,
+    )
+    general_cfg = config.General(
+        log_level=log_level,
+        log_file=log_file,
+        quiet=effective_quiet,
+        clipboard=not json_output,
+    )
+
+    corrected_text = asyncio.run(
+        _async_autocorrect(
+            text=text,
+            provider_cfg=provider_cfg,
+            ollama_cfg=ollama_cfg,
+            openai_llm_cfg=openai_llm_cfg,
+            gemini_llm_cfg=gemini_llm_cfg,
+            general_cfg=general_cfg,
+        ),
+    )
+    if json_output:
+        print(json.dumps({"corrected_text": corrected_text}))