devorch 0.3.0__tar.gz → 0.4.0__tar.gz

devorch-0.4.0/DEVORCH.md

@@ -0,0 +1,48 @@
+# DevOrch - AI Coding Assistant CLI
+
+## Overview
+DevOrch is an enterprise-ready, multi-provider AI coding assistant CLI with support for 13+ LLM providers, interactive REPL, persistent sessions, tool execution, MCP servers, and background terminal processes.
+
+## Tech Stack
+- **Language**: Python 3.10+ (tested on Python 3.10-3.13)
+- **CLI Framework**: Typer, Questionary, Prompt Toolkit, Rich
+- **AI / SDKs**: OpenAI, Anthropic, Google GenAI SDK, HTTPX
+- **Data Validation**: Pydantic v2
+- **Testing & Quality**: Pytest, Pytest-asyncio, Ruff, Mypy
+
+## Architecture & Project Structure
+- `cli/`: CLI entrypoints, Typer commands (`ask`, `edit`, `run`, `init`), slash command handlers, banners, and REPL.
+- `core/`: Agent orchestration engine:
+  - `agent.py`: Main agentic loop, tool calling, execution loop, plan mode approval.
+  - `context.py`: Token estimation, context compaction, tool output pruning, token usage tracking.
+  - `loop_detector.py`: Duplicate call detection, oscillation detection, error thrashing prevention, turn advisories.
+  - `project_context.py`: Auto-discovers and loads repository instructions (`DEVORCH.md`, `CLAUDE.md`, `AGENTS.md`).
+  - `executor.py`: Tool permission enforcement and invocation.
+  - `planner.py`: Message preparation and system prompt injection.
+  - `sessions.py`: Session persistence, continuation, and summarization.
+  - `modes.py`: Agent execution modes (`ASK`, `AUTO`, `PLAN`).
+  - `memory.py` & `skills.py`: Persistent memory and skill definitions.
+- `tools/`: Built-in tools:
+  - `filesystem.py`: File reading with pagination, writing, directory listing.
+  - `edit.py`: Surgical find/replace, line-based editing, diff generation.
+  - `grep.py`: Fast regex pattern search across text files.
+  - `search.py`: Glob-based file and directory search.
+  - `shell.py` & `terminal_session.py`: Shell execution and persistent terminal sessions.
+- `providers/`: Unified LLM provider adapters (`openai`, `anthropic`, `gemini`, `groq`, `deepseek`, etc.).
+- `schemas/`: Dataclasses for `Message`, `ToolCall`, `LLMResponse`, `TokenUsage`.
+
+## Common Commands
+- **Run Tests**: `pytest` or `python -m pytest`
+- **Run Single Test**: `pytest tests/test_context.py`
+- **Lint**: `ruff check .`
+- **Format**: `ruff format .`
+- **Type Check**: `mypy .`
+- **Run DevOrch Locally**: `devorch` or `python -m cli.main`
+
+## Coding Standards & Minimal Token Guidelines
+1. **Find Before Reading**: Always use `search` (to find file paths) or `grep` (to find symbols/definitions) before reading file contents.
+2. **Inspect Surgically**: Use `filesystem` with `action="read_lines"` or specific line ranges rather than dumping huge files.
+3. **Edit Surgically**: Use `edit` with `action="replace"` or `action="replace_lines"` to generate clean unified diffs. Avoid full file rewrites whenever possible.
+4. **Prevent Loops**: Never repeat identical tool calls with the same arguments. If a tool fails or finds nothing, alter the query or check file paths.
+5. **Verify Changes**: After modifying code, always run the relevant pytest test suite to verify correctness before reporting completion.
+6. **Preserve Compatibility**: Keep type hints compatible with Python 3.10+ (use `|` for unions or `Optional`/`Union`).

{devorch-0.3.0 → devorch-0.4.0}/PKG-INFO

@@ -1,6 +1,6 @@
-Metadata-Version: 2.4
+Metadata-Version: 2.5
 Name: devorch
-Version: 0.3.0
+Version: 0.4.0
 Summary: Multi-provider AI coding assistant CLI with 13+ providers
 Project-URL: Homepage, https://github.com/Amanbig/DevOrch
 Project-URL: Repository, https://github.com/Amanbig/DevOrch
@@ -76,7 +76,9 @@ DevOrch gives you a coding assistant in your terminal that can execute shell com
 
 - **Provider freedom** — Switch between OpenAI, Anthropic, Gemini, Mistral, Groq, and 8 more providers (including local models) with a single command. No vendor lock-in.
 - **Actually does things** — Runs shell commands, edits files, manages background processes, searches the web. Not just a chatbot.
-- **Remembers you** — Persistent memory system stores your preferences, project context, and feedback across conversations.
+- **Cross-session project memory** — Per-project persistent memory system automatically recalls architecture decisions, coding preferences, and recent session milestones across sessions.
+- **Loop & thrashing safeguards** — Advanced safety layer detects duplicate calls, tool oscillation, and error loops, preventing runaway token drain.
+- **Full token & speed transparency** — Real-time prompt/completion token tracking, context compaction metrics, and generation speed (`tok/s`) across 13 providers.
 - **Extensible** — Add custom skills as YAML files, connect MCP servers for additional tools, configure permissions per-tool.
 
 ## Quick Start
@@ -142,18 +144,27 @@ DevOrch can act on your system, not just talk about it:
 | `memory` | Persistent memory across conversations |
 | `websearch` / `webfetch` | Search the web, fetch URLs |
 
-### Memory System
-
-DevOrch remembers context across conversations:
-
-```
-/remember I prefer TypeScript over JavaScript
-/remember This project uses PostgreSQL, not MySQL
-/memory                    # View all saved memories
-/forget                    # Delete a memory interactively
+### Memory & Project Context
+ 
+DevOrch maintains persistent memory so you never have to repeat context:
+ 
+**1. Per-Project Memory (Cross-Session Recall)**
+Stored automatically per repository in `~/.devorch/projects/<project_id>/`:
+```bash
+/memory                              # View active architectural decisions & session history
+/memory add Use uv for package management   # Record an architectural decision
+/memory pref Prefer clean async/await       # Record a coding style preference
+/memory clear                        # Clear project memory for this repository
 ```
+*Accomplishments are automatically extracted and summarized on session exit with zero raw command leakage.*
 
-Memory types: **user** (preferences), **feedback** (corrections), **project** (context), **reference** (external links).
+**2. Global Memory (Across All Projects)**
+Stored in `~/.devorch/memory/`:
+```bash
+/remember I prefer tabs over spaces
+/remember Use ruff for Python linting
+/forget                              # Interactively choose a memory to remove
+```
 
 ### Skills
 
@@ -285,6 +296,19 @@ devorch edit app/models.py "add created_at field" --mcp sqlite
 
 # List available skills
 devorch skills
+
+# Browse providers and models directly from terminal
+devorch providers               # View all 13 providers & configuration status
+devorch models                  # View models for active provider
+devorch models anthropic        # View models for a specific provider
+
+# Inspect or manage project memory
+devorch memory                  # View project decisions and recent accomplishments
+devorch memory add "Use pytest" # Record a decision from the terminal
+devorch memory clear            # Clear project memory
+
+# Repository initialization
+devorch init                    # Generate or update DEVORCH.md repository instructions
 ```
 
 All non-interactive commands support `--provider`, `--model`, `--mode`, `--mcp`, and `--no-mcp`.
@@ -293,33 +317,38 @@ All non-interactive commands support `--provider`, `--model`, `--mode`, `--mcp`,
 
 | Command | Description |
 |---------|-------------|
-| `/help` | Show all commands |
-| `/models` | Browse and switch models (interactive) |
-| `/model <name>` | Switch model (partial match supported) |
-| `/providers` | Browse and switch providers (interactive) |
-| `/provider <name>` | Switch provider directly |
-| `/mode` | Switch mode (Plan/Auto/Ask) |
-| `/status` | Show current config |
-| `/auth [provider]` | Set/update API key |
-| `/memory` | Show saved memories |
-| `/remember <text>` | Save to memory |
-| `/forget` | Delete a memory |
+| `/help` | Show categorized commands and shortcuts |
+| `/mode [plan\|auto\|ask]` | Show or switch execution mode |
+| `/plan` `/auto` `/ask` | Quick switch execution mode |
+| `/models` | Search and browse models with 15/page pagination |
+| `/model <name>` | Switch model directly or launch search |
+| `/providers` | Search and browse providers with interactive pagination |
+| `/provider <name>` | Switch provider directly or launch search |
+| `/tokens` | Show session token usage, prompt/completion split, and costs |
+| `/copy` | Copy last assistant response directly to clipboard |
+| `/paste` | Enter multi-line paste mode for large text or code (`Alt+Enter` also inserts newlines) |
+| `/status` | Show current provider, model, execution mode, and loaded context |
+| `/auth [provider]` | Set or update API key for any provider |
+| `/memory` | Show project memory (`/memory add <dec>`, `/memory pref <p>`, `/memory clear`) |
+| `/remember <text>` | Save a note or convention to global memory |
+| `/forget` | Delete a global memory interactively or by query |
+| `/init` | Generate or update a `DEVORCH.md` project context file |
 | `/skills` | List available skills |
-| `/skill <name>` | Run a skill |
+| `/skill <name>` | Run a skill directly |
 | `/commit` `/review` `/test` `/fix` `/explain` `/simplify` | Skill shortcuts |
-| `/session` | Session info |
-| `/history` | Conversation history |
-| `/clear` | Clear history |
-| `/compact` | Summarize history |
-| `/save` | Save to file |
-| `/undo` | Undo last message |
+| `/session` | Current session ID, model, and message count |
+| `/history` | Full conversation history in this session |
+| `/clear` | Clear history (saves accomplishments to project memory) |
+| `/compact` | Summarize and compact conversation history to save tokens |
+| `/save` | Save conversation history to a file |
+| `/undo` | Undo last message and agent turn |
 | `/mcp` | Show MCP server status |
 | `/mcp add <name> <cmd> [args]` | Connect a new MCP server mid-session |
 | `/mcp start <name>` | Reconnect a server from config |
 | `/mcp stop <name>` | Disconnect a server and remove its tools |
-| `/config` | Show configuration |
-| `/permissions` | Show permissions |
-| `/tasks` | Show task list |
+| `/config` | Show configuration settings |
+| `/permissions` | Show and manage tool permission settings |
+| `/tasks` | Show multi-step task list and execution progress |
 
 ## Configuration
 

{devorch-0.3.0 → devorch-0.4.0}/README.md

@@ -34,7 +34,9 @@ DevOrch gives you a coding assistant in your terminal that can execute shell com
 
 - **Provider freedom** — Switch between OpenAI, Anthropic, Gemini, Mistral, Groq, and 8 more providers (including local models) with a single command. No vendor lock-in.
 - **Actually does things** — Runs shell commands, edits files, manages background processes, searches the web. Not just a chatbot.
-- **Remembers you** — Persistent memory system stores your preferences, project context, and feedback across conversations.
+- **Cross-session project memory** — Per-project persistent memory system automatically recalls architecture decisions, coding preferences, and recent session milestones across sessions.
+- **Loop & thrashing safeguards** — Advanced safety layer detects duplicate calls, tool oscillation, and error loops, preventing runaway token drain.
+- **Full token & speed transparency** — Real-time prompt/completion token tracking, context compaction metrics, and generation speed (`tok/s`) across 13 providers.
 - **Extensible** — Add custom skills as YAML files, connect MCP servers for additional tools, configure permissions per-tool.
 
 ## Quick Start
@@ -100,18 +102,27 @@ DevOrch can act on your system, not just talk about it:
 | `memory` | Persistent memory across conversations |
 | `websearch` / `webfetch` | Search the web, fetch URLs |
 
-### Memory System
-
-DevOrch remembers context across conversations:
-
-```
-/remember I prefer TypeScript over JavaScript
-/remember This project uses PostgreSQL, not MySQL
-/memory                    # View all saved memories
-/forget                    # Delete a memory interactively
+### Memory & Project Context
+ 
+DevOrch maintains persistent memory so you never have to repeat context:
+ 
+**1. Per-Project Memory (Cross-Session Recall)**
+Stored automatically per repository in `~/.devorch/projects/<project_id>/`:
+```bash
+/memory                              # View active architectural decisions & session history
+/memory add Use uv for package management   # Record an architectural decision
+/memory pref Prefer clean async/await       # Record a coding style preference
+/memory clear                        # Clear project memory for this repository
 ```
+*Accomplishments are automatically extracted and summarized on session exit with zero raw command leakage.*
 
-Memory types: **user** (preferences), **feedback** (corrections), **project** (context), **reference** (external links).
+**2. Global Memory (Across All Projects)**
+Stored in `~/.devorch/memory/`:
+```bash
+/remember I prefer tabs over spaces
+/remember Use ruff for Python linting
+/forget                              # Interactively choose a memory to remove
+```
 
 ### Skills
 
@@ -243,6 +254,19 @@ devorch edit app/models.py "add created_at field" --mcp sqlite
 
 # List available skills
 devorch skills
+
+# Browse providers and models directly from terminal
+devorch providers               # View all 13 providers & configuration status
+devorch models                  # View models for active provider
+devorch models anthropic        # View models for a specific provider
+
+# Inspect or manage project memory
+devorch memory                  # View project decisions and recent accomplishments
+devorch memory add "Use pytest" # Record a decision from the terminal
+devorch memory clear            # Clear project memory
+
+# Repository initialization
+devorch init                    # Generate or update DEVORCH.md repository instructions
 ```
 
 All non-interactive commands support `--provider`, `--model`, `--mode`, `--mcp`, and `--no-mcp`.
@@ -251,33 +275,38 @@ All non-interactive commands support `--provider`, `--model`, `--mode`, `--mcp`,
 
 | Command | Description |
 |---------|-------------|
-| `/help` | Show all commands |
-| `/models` | Browse and switch models (interactive) |
-| `/model <name>` | Switch model (partial match supported) |
-| `/providers` | Browse and switch providers (interactive) |
-| `/provider <name>` | Switch provider directly |
-| `/mode` | Switch mode (Plan/Auto/Ask) |
-| `/status` | Show current config |
-| `/auth [provider]` | Set/update API key |
-| `/memory` | Show saved memories |
-| `/remember <text>` | Save to memory |
-| `/forget` | Delete a memory |
+| `/help` | Show categorized commands and shortcuts |
+| `/mode [plan\|auto\|ask]` | Show or switch execution mode |
+| `/plan` `/auto` `/ask` | Quick switch execution mode |
+| `/models` | Search and browse models with 15/page pagination |
+| `/model <name>` | Switch model directly or launch search |
+| `/providers` | Search and browse providers with interactive pagination |
+| `/provider <name>` | Switch provider directly or launch search |
+| `/tokens` | Show session token usage, prompt/completion split, and costs |
+| `/copy` | Copy last assistant response directly to clipboard |
+| `/paste` | Enter multi-line paste mode for large text or code (`Alt+Enter` also inserts newlines) |
+| `/status` | Show current provider, model, execution mode, and loaded context |
+| `/auth [provider]` | Set or update API key for any provider |
+| `/memory` | Show project memory (`/memory add <dec>`, `/memory pref <p>`, `/memory clear`) |
+| `/remember <text>` | Save a note or convention to global memory |
+| `/forget` | Delete a global memory interactively or by query |
+| `/init` | Generate or update a `DEVORCH.md` project context file |
 | `/skills` | List available skills |
-| `/skill <name>` | Run a skill |
+| `/skill <name>` | Run a skill directly |
 | `/commit` `/review` `/test` `/fix` `/explain` `/simplify` | Skill shortcuts |
-| `/session` | Session info |
-| `/history` | Conversation history |
-| `/clear` | Clear history |
-| `/compact` | Summarize history |
-| `/save` | Save to file |
-| `/undo` | Undo last message |
+| `/session` | Current session ID, model, and message count |
+| `/history` | Full conversation history in this session |
+| `/clear` | Clear history (saves accomplishments to project memory) |
+| `/compact` | Summarize and compact conversation history to save tokens |
+| `/save` | Save conversation history to a file |
+| `/undo` | Undo last message and agent turn |
 | `/mcp` | Show MCP server status |
 | `/mcp add <name> <cmd> [args]` | Connect a new MCP server mid-session |
 | `/mcp start <name>` | Reconnect a server from config |
 | `/mcp stop <name>` | Disconnect a server and remove its tools |
-| `/config` | Show configuration |
-| `/permissions` | Show permissions |
-| `/tasks` | Show task list |
+| `/config` | Show configuration settings |
+| `/permissions` | Show and manage tool permission settings |
+| `/tasks` | Show multi-step task list and execution progress |
 
 ## Configuration
 

{devorch-0.3.0 → devorch-0.4.0}/cli/commands/_shared.py

@@ -4,6 +4,9 @@ Shared utilities for DevOrch CLI commands.
 Exports: console, SYSTEM_PROMPT, SimplePlanner, create_provider, build_tools, resolve_mode
 """
 
+import os
+import platform
+
 import typer
 
 from config.settings import Settings
@@ -32,10 +35,10 @@ SYSTEM_PROMPT = """You are DevOrch, an AI coding assistant with access to tools
 
 IMPORTANT: You have the following tools available and MUST use them to help the user:
 
-1. **shell** - Execute shell commands (bash/powershell). Use this to:
-   - Run commands like `npm install`, `git clone`, `git status`, etc.
-   - Navigate directories, create files, run scripts
-   - Any short-lived terminal command that returns output
+1. **shell** - Execute shell commands (builds, tests, git, package managers). Use this to:
+   - Run CLI tools: `npm`, `pip`, `uv`, `pytest`, `cargo`, `git`
+   - Run project test suites, linters, and compilation scripts
+   - Do NOT use shell to inspect directory contents or read files — use `filesystem` instead!
 
 2. **terminal_session** — The PRIMARY tool for all terminal/process needs:
    - `start` — launch a command in a managed session. Defaults to 'bash' if no command.
@@ -112,12 +115,27 @@ RULES:
 - When the user corrects you or gives feedback, save it to memory for future conversations
 - When you learn about the user's role, preferences, or project context, save it to memory
 
-When executing shell commands, use the shell tool with the command to run."""
+- When executing shell commands, use the shell tool with the command to run.
+
+EFFICIENCY & SURGICAL WORKFLOW (MINIMIZE TOKENS):
+- **Find before reading**: Use `search` (to locate files) or `grep` (to find specific functions/classes/patterns) instead of listing directory trees or reading arbitrary files.
+- **Inspect surgically**: Use `filesystem` with `action="read_lines"` or bounded `max_lines` to inspect specific sections. Do NOT dump entire huge files into context when only a small section is needed.
+- **Edit surgically**: Use `edit` with `action="replace"` or `action="replace_lines"` to make precise changes.
+- **Verify without waste**: After making an edit, do NOT re-read the entire file if the diff confirmed the change. Verify with test or lint commands when applicable.
+- **Avoid loops**: Never call the exact same tool with the exact same arguments repeatedly. If an edit or search fails, re-examine the target lines or change the query before retrying."""
 
 
 class SimplePlanner(Planner):
-    def __init__(self, memory_context: str = "", tools: list | None = None):
+    def __init__(
+        self,
+        memory_context: str = "",
+        project_context: str = "",
+        project_memory_context: str = "",
+        tools: list | None = None,
+    ):
         self.memory_context = memory_context
+        self.project_context = project_context
+        self.project_memory_context = project_memory_context
         self._tools = tools or []
 
     def update_tools(self, tools: list) -> None:
@@ -127,6 +145,44 @@ class SimplePlanner(Planner):
     def plan(self, history: list[Message]) -> list[Message]:
         prompt = SYSTEM_PROMPT
 
+        # Append runtime environment & host OS shell guidance
+        os_name = platform.system()
+        os_release = platform.release()
+        os_machine = platform.machine()
+        cwd = os.getcwd()
+
+        if os_name == "Windows":
+            shell_info = (
+                f"\n\nENVIRONMENT & SYSTEM CONTEXT:\n"
+                f"- Host OS: Windows ({os_release}, {os_machine})\n"
+                f"- Current Working Directory: {cwd}\n"
+                f"- Shell: Windows cmd.exe / PowerShell\n"
+                f"- IMPORTANT SHELL RULES FOR WINDOWS:\n"
+                f"  * Do NOT use Unix shell commands like `pwd`, `ls`, `ls -la`, `cat`, `touch`, `rm -rf`, `grep`, or `export`.\n"
+                f"  * To inspect directory contents: use the `filesystem` tool with `action='list'`, `path='.'`.\n"
+                f"  * To read files: use the `filesystem` tool with `action='read'`.\n"
+                f"  * To search files: use `search` (glob) or `grep` (content search).\n"
+                f"  * For builds & CLI tools, use direct commands: `pip`, `uv`, `npm`, `git`, `python`, `pytest`.\n"
+                f"  * For Windows file paths in shell commands, use backslashes `\\` or wrap paths in quotes."
+            )
+        else:
+            shell_info = (
+                f"\n\nENVIRONMENT & SYSTEM CONTEXT:\n"
+                f"- Host OS: {os_name} ({os_release}, {os_machine})\n"
+                f"- Current Working Directory: {cwd}\n"
+                f"- Shell: bash/sh\n"
+                f"- Prefer using `filesystem` tool for reading and listing files to minimize token usage."
+            )
+        prompt += shell_info
+
+        # Append project rules/guidelines (from DEVORCH.md or CLAUDE.md)
+        if self.project_context:
+            prompt += "\n\n" + self.project_context
+
+        # Append persistent per-project memory (cross-session decisions & learnings)
+        if self.project_memory_context:
+            prompt += "\n\n" + self.project_memory_context
+
         # Append a live tool summary so the LLM knows exactly what's loaded,
         # including any MCP tools added after startup.
         if self._tools:
@@ -242,7 +298,14 @@ def resolve_mode(mode_str: str | None, default: AgentMode = AgentMode.AUTO) -> A
         raise typer.Exit(1) from err
 
 
-def build_agent(llm, tools: list, memory_ctx: str, agent_mode: AgentMode) -> "Agent":
+def build_agent(
+    llm,
+    tools: list,
+    memory_ctx: str,
+    agent_mode: AgentMode,
+    project_ctx: str = "",
+    project_memory_ctx: str = "",
+) -> "Agent":
     """Construct a configured Agent ready to run.
 
     AgentTool is injected after construction since it needs the provider reference.
@@ -251,7 +314,12 @@ def build_agent(llm, tools: list, memory_ctx: str, agent_mode: AgentMode) -> "Ag
     """
     mode_manager = ModeManager(default_mode=agent_mode)
     executor = ToolExecutor(tools=tools, mode_manager=mode_manager)
-    planner = SimplePlanner(memory_context=memory_ctx, tools=tools)
+    planner = SimplePlanner(
+        memory_context=memory_ctx,
+        project_context=project_ctx,
+        project_memory_context=project_memory_ctx,
+        tools=tools,
+    )
 
     agent = Agent(
         provider=llm,

{devorch-0.3.0 → devorch-0.4.0}/cli/commands/ask.py

@@ -6,6 +6,7 @@ from cli.commands._shared import build_agent, build_tools, console, create_provi
 from config.settings import Settings
 from core.memory import MemoryManager
 from core.modes import AgentMode
+from core.project_context import ProjectContextLoader
 from core.skills import SkillManager
 from utils.logger import print_error, print_panel
 
@@ -41,7 +42,9 @@ def ask(
     tools = build_tools(settings, no_mcp=no_mcp, mcp_only=mcp or None)
     agent_mode = resolve_mode(mode, default=AgentMode.AUTO)
     memory_ctx = MemoryManager().get_context_prompt()
-    agent = build_agent(llm, tools, memory_ctx, agent_mode)
+    project_ctx = ProjectContextLoader().load()
+    project_prompt = project_ctx.to_system_prompt_block() if project_ctx else ""
+    agent = build_agent(llm, tools, memory_ctx, agent_mode, project_ctx=project_prompt)
 
     skill_manager = SkillManager()
     if skill:

{devorch-0.3.0 → devorch-0.4.0}/cli/commands/run.py

@@ -5,6 +5,7 @@ import typer
 from cli.commands._shared import build_agent, build_tools, console, create_provider, resolve_mode
 from config.settings import Settings
 from core.memory import MemoryManager
+from core.project_context import ProjectContextLoader
 from core.skills import SkillManager
 from utils.logger import print_error, print_panel
 
@@ -49,7 +50,9 @@ def run(
     tools = build_tools(settings, no_mcp=no_mcp, mcp_only=mcp or None)
     agent_mode = resolve_mode(mode)
     memory_ctx = MemoryManager().get_context_prompt()
-    agent = build_agent(llm, tools, memory_ctx, agent_mode)
+    project_ctx = ProjectContextLoader().load()
+    project_prompt = project_ctx.to_system_prompt_block() if project_ctx else ""
+    agent = build_agent(llm, tools, memory_ctx, agent_mode, project_ctx=project_prompt)
 
     console.print(f"[dim]Skill: {skill_name} | {llm.name}/{llm.model}[/dim]")
     try:

{devorch-0.3.0 → devorch-0.4.0}/cli/constants.py

@@ -17,7 +17,7 @@ console = get_console()
 
 # ── Version ──────────────────────────────────────────────────────────────────
 
-VERSION = "0.3.0"
+VERSION = "0.4.0"
 
 # ── Banners ───────────────────────────────────────────────────────────────────
 
@@ -33,32 +33,36 @@ BANNER_SMALL = "[bold cyan]DevOrch[/bold cyan]"
 # automatically.
 
 SLASH_COMMANDS: dict[str, str] = {
-    "/help": "Show available commands",
-    "/mode": "Show or change mode (plan/auto/ask)",
-    "/plan": "Switch to plan mode",
-    "/auto": "Switch to auto mode",
-    "/ask": "Switch to ask mode (default)",
-    "/clear": "Clear conversation history",
-    "/session": "Show current session info",
-    "/config": "Show configuration settings",
-    "/permissions": "Show permission settings",
-    "/compact": "Summarize and compact history",
-    "/models": "Browse and switch models (interactive)",
-    "/model": "Switch model (/model <name> or interactive)",
-    "/providers": "Browse and switch providers (interactive)",
-    "/provider": "Switch provider (/provider <name> or interactive)",
-    "/history": "Show conversation history",
-    "/undo": "Undo last message",
-    "/save": "Save conversation to file",
-    "/status": "Show current provider, model, and mode",
-    "/tasks": "Show current task list",
-    "/memory": "Show saved memories",
-    "/remember": "Save something to memory",
-    "/forget": "Delete a memory",
-    "/skills": "List available skills",
-    "/skill": "Run a skill (e.g. /skill commit)",
-    "/mcp": "Show MCP servers | add/stop/start servers inline",
-    "/auth": "Set or update API key for current/specified provider",
+    "/help": "Show categorized help and command shortcuts",
+    "/mode": "Show or switch mode: /mode [plan|auto|ask]",
+    "/plan": "Switch to plan mode (requires approval before tool execution)",
+    "/auto": "Switch to auto mode (executes approved tools autonomously)",
+    "/ask": "Switch to ask mode (default: read and suggest only)",
+    "/clear": "Clear conversation history (saves accomplishments to project memory)",
+    "/session": "Show current session ID, model, and message count",
+    "/config": "Show configuration settings and provider status",
+    "/permissions": "Show and manage tool permission settings",
+    "/compact": "Summarize and compact conversation history to save tokens",
+    "/models": "Search, browse, and switch models (interactive 15/page pagination)",
+    "/model": "Switch model directly (/model <name>) or open interactive search",
+    "/providers": "Search, browse, and switch providers (interactive pagination)",
+    "/provider": "Switch provider directly (/provider <name>) or open interactive search",
+    "/history": "Show full conversation history in this session",
+    "/undo": "Undo last message and agent turn",
+    "/save": "Save conversation history to a file",
+    "/status": "Show active provider, model, execution mode, and loaded context",
+    "/tasks": "Show active multi-step task list and execution progress",
+    "/memory": "Show project memory (/memory add <dec>, /memory pref <p>, /memory clear)",
+    "/remember": "Save something to global memory (/remember <note>)",
+    "/forget": "Delete a memory (/forget or /forget <name>)",
+    "/skills": "List available built-in and custom skills",
+    "/skill": "Run a skill directly (/skill <name>)",
+    "/mcp": "Show MCP servers (/mcp add, /mcp stop, /mcp start)",
+    "/auth": "Set or update API key for current or specified provider",
+    "/tokens": "Show session token usage, prompt/completion split, and costs",
+    "/copy": "Copy the last assistant response to clipboard",
+    "/paste": "Enter multi-line paste mode to input large text or code",
+    "/init": "Generate or update a DEVORCH.md project context file",
 }
 
 # ── Questionary style (provider/model selection prompts) ─────────────────────
@@ -83,6 +87,7 @@ QUESTIONARY_STYLE = QStyle(
 PROMPT_STYLE = Style.from_dict(
     {
         "prompt": "#55cc55 bold",
+        "prompt-mode": "#55aaff bold",
         "prompt-arrow": "#55cc55 bold",
         "": "#ffffff bold",
         "command": "#66ccff bold",