vamp-cli 0.1.4__tar.gz → 0.1.6__tar.gz

{vamp_cli-0.1.4/src/vamp_cli.egg-info → vamp_cli-0.1.6}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: vamp-cli
-Version: 0.1.4
+Version: 0.1.6
 Summary: VAMP-CLI is an interactive coding agent for working in a local repository. It combines a planner, a coder, model fallbacks, workspace-aware file tools, and a Docker-backed command sandbox in one terminal application. The agent can inspect and edit files, search a codebase, run commands, maintain a task list, and ask for confirmation before executing potentially dangerous operations.
 Author-email: Satyam Kumar <satkr2208@gmail.com>
 License: MIT License
@@ -63,7 +63,7 @@ The agent can inspect and edit files, search a codebase, run commands, maintain
 
 ## Highlights
 
-- Interactive terminal UI launched with the `vamp` command.
+- Interactive terminal UI launched with the `vamp` command, powered by **[Rich](https://github.com/Textualize/rich)** for beautiful formatting, colors, and progress indicators.
 - Two operating modes:
   - **Normal mode**: work directly with the coding agent.
   - **Plan mode**: have the planner produce a plan and approve it before implementation.
@@ -79,11 +79,23 @@ The agent can inspect and edit files, search a codebase, run commands, maintain
 - **Branch switching**: view and switch between conversation branches within a session.
 - **Conversation compaction**: automatically or manually summarize early turns to stay within model context limits.
 
+## Demo
+
+### Interactive Terminal Experience
+
+Below is a demonstration of the VAMP-CLI interface in action. The UI utilizes the **Rich** library to provide syntax highlighting, styled tables, and clear visual feedback during agent operations.
+
+![VAMP-CLI Demo GIF](./assets/demo.gif)
+
+### Interface Screenshot
+
+![VAMP-CLI Interface Screenshot](./assets/screenshot.png)
+
 ## Requirements
 
 - Python 3.11 or newer
 - Docker Engine or Docker Desktop, running and accessible to the current user
-- An API key for each model configured in `src/vamp_cli/config/config.yaml`
+- An API key env var for each model configured in `~/.vamp/config.yaml` (it gets created in first run).
 
 The first run automatically builds the `vamp-sandbox:latest` image from [`src/vamp_cli/sandbox/Dockerfile`](src/vamp_cli/sandbox/Dockerfile) when it is not already available locally. The image includes Python, Node.js, Git, `ripgrep`, `fd`, and common build tools.
 
@@ -91,13 +103,26 @@ The first run automatically builds the `vamp-sandbox:latest` image from [`src/va
 
 Create and activate a virtual environment, then install the project (all dependencies will be installed itself):
 
+Virtual Env:
+
 ```bash
 python3.12 -m venv .venv
 source .venv/bin/activate
 python -m pip install --upgrade pip
+```
+
+Package:
+
+```bash
 python -m pip install vamp-cli 
 ```
 
+For development:
+
+```bash
+python -m pip install -e .
+```
+
 Confirm that Docker is available before starting the agent:
 
 ```bash
@@ -106,7 +131,7 @@ docker info
 
 ## Configuration
 
-Model and profile configuration lives in [`src/vamp_cli/config/config.yaml`](src/vamp_cli/config/config.yaml). API keys are read from environment variables named by each model's `api_key_env` field; keys are not read from YAML values directly.
+Model and profile configuration lives in [`/src/vamp_cli/config/default_config.yaml`](/src/vamp_cli/config/default_config.yaml)(this is used to create `config.yaml` for first run). After first run `~/.vamp/config.yaml` is created and vamp-cli reads from this always. API keys are read from environment variables named by each model's `api_key_env` field; keys are not read from YAML values directly.
 
 The checked-in default configuration defines `qwen` and `openai`, so both variables must be present unless you edit the configuration to remove or replace one of those models:
 
@@ -136,6 +161,7 @@ You:
 | Argument | Action |
 | --- | --- |
 | `--resume <id>` or `-r <id>` | Resume a previous session by its ID |
+| `--previous` or `-p` | Resume last session |
 | `--list` or `-l` | List all past sessions |
 
 Examples:
@@ -161,6 +187,7 @@ vamp --resume 01a145b4-5103-7014-97da-d5ca9c871d68
 | `/compact` | Summarize early conversation turns to reduce context usage |
 | `/compact <N>` | Compact keeping the last N turns intact (default: 4) |
 | `/config` | Change configurations of models, coder or planner profiles, and tool access. |
+| `@<file-path>` | Use `@` to select a file and ask agent about that file. |
 
 When the agent requests approval, answer `yes` to continue or `no` and provide a reason to deny the action. Approvals are used for plan implementation and security-sensitive tool calls such as destructive commands, package installation, or network access.
 

{vamp_cli-0.1.4 → vamp_cli-0.1.6}/README.md

@@ -8,7 +8,7 @@ The agent can inspect and edit files, search a codebase, run commands, maintain
 
 ## Highlights
 
-- Interactive terminal UI launched with the `vamp` command.
+- Interactive terminal UI launched with the `vamp` command, powered by **[Rich](https://github.com/Textualize/rich)** for beautiful formatting, colors, and progress indicators.
 - Two operating modes:
   - **Normal mode**: work directly with the coding agent.
   - **Plan mode**: have the planner produce a plan and approve it before implementation.
@@ -24,11 +24,23 @@ The agent can inspect and edit files, search a codebase, run commands, maintain
 - **Branch switching**: view and switch between conversation branches within a session.
 - **Conversation compaction**: automatically or manually summarize early turns to stay within model context limits.
 
+## Demo
+
+### Interactive Terminal Experience
+
+Below is a demonstration of the VAMP-CLI interface in action. The UI utilizes the **Rich** library to provide syntax highlighting, styled tables, and clear visual feedback during agent operations.
+
+![VAMP-CLI Demo GIF](./assets/demo.gif)
+
+### Interface Screenshot
+
+![VAMP-CLI Interface Screenshot](./assets/screenshot.png)
+
 ## Requirements
 
 - Python 3.11 or newer
 - Docker Engine or Docker Desktop, running and accessible to the current user
-- An API key for each model configured in `src/vamp_cli/config/config.yaml`
+- An API key env var for each model configured in `~/.vamp/config.yaml` (it gets created in first run).
 
 The first run automatically builds the `vamp-sandbox:latest` image from [`src/vamp_cli/sandbox/Dockerfile`](src/vamp_cli/sandbox/Dockerfile) when it is not already available locally. The image includes Python, Node.js, Git, `ripgrep`, `fd`, and common build tools.
 
@@ -36,13 +48,26 @@ The first run automatically builds the `vamp-sandbox:latest` image from [`src/va
 
 Create and activate a virtual environment, then install the project (all dependencies will be installed itself):
 
+Virtual Env:
+
 ```bash
 python3.12 -m venv .venv
 source .venv/bin/activate
 python -m pip install --upgrade pip
+```
+
+Package:
+
+```bash
 python -m pip install vamp-cli 
 ```
 
+For development:
+
+```bash
+python -m pip install -e .
+```
+
 Confirm that Docker is available before starting the agent:
 
 ```bash
@@ -51,7 +76,7 @@ docker info
 
 ## Configuration
 
-Model and profile configuration lives in [`src/vamp_cli/config/config.yaml`](src/vamp_cli/config/config.yaml). API keys are read from environment variables named by each model's `api_key_env` field; keys are not read from YAML values directly.
+Model and profile configuration lives in [`/src/vamp_cli/config/default_config.yaml`](/src/vamp_cli/config/default_config.yaml)(this is used to create `config.yaml` for first run). After first run `~/.vamp/config.yaml` is created and vamp-cli reads from this always. API keys are read from environment variables named by each model's `api_key_env` field; keys are not read from YAML values directly.
 
 The checked-in default configuration defines `qwen` and `openai`, so both variables must be present unless you edit the configuration to remove or replace one of those models:
 
@@ -81,6 +106,7 @@ You:
 | Argument | Action |
 | --- | --- |
 | `--resume <id>` or `-r <id>` | Resume a previous session by its ID |
+| `--previous` or `-p` | Resume last session |
 | `--list` or `-l` | List all past sessions |
 
 Examples:
@@ -106,6 +132,7 @@ vamp --resume 01a145b4-5103-7014-97da-d5ca9c871d68
 | `/compact` | Summarize early conversation turns to reduce context usage |
 | `/compact <N>` | Compact keeping the last N turns intact (default: 4) |
 | `/config` | Change configurations of models, coder or planner profiles, and tool access. |
+| `@<file-path>` | Use `@` to select a file and ask agent about that file. |
 
 When the agent requests approval, answer `yes` to continue or `no` and provide a reason to deny the action. Approvals are used for plan implementation and security-sensitive tool calls such as destructive commands, package installation, or network access.
 

{vamp_cli-0.1.4 → vamp_cli-0.1.6}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "vamp-cli"
-version = "0.1.4"
+version = "0.1.6"
 authors = [
   { name = "Satyam Kumar", email = "satkr2208@gmail.com" }
 ]

{vamp_cli-0.1.4 → vamp_cli-0.1.6}/src/vamp_cli/agent/nodes.py

@@ -9,7 +9,7 @@ from prompt_toolkit.styles import Style
 from langgraph.graph import END
 from langgraph.types import interrupt, Command
 from langchain_core.runnables import Runnable, RunnableConfig
-from langchain_core.messages import ToolMessage, SystemMessage, AIMessage, BaseMessage
+from langchain_core.messages import ToolMessage, SystemMessage, AIMessage, BaseMessage, HumanMessage
 
 from vamp_cli.tools.registry import ToolRegistry
 from vamp_cli.agent.state import AgentState, TodoItem
@@ -36,24 +36,24 @@ def create_planner_node(model: Runnable):
             None
         )
         
-        if response.tool_calls:
-            print_formatted_text(
-                FormattedText([(
-                    "class:dimmed",
-                    "\n\n[planner] calling:"
-                )]),
-                style=_dim_style,
-                end="\n"
-            )
-            for tc in response.tool_calls:
-                print_formatted_text(
-                    FormattedText([(
-                        "class:dimmed", 
-                        f">> {str(tc['name'])}({str(tc['args'])})"
-                    )]),
-                    style=_dim_style,
-                    end="\n"
-                )
+        # if response.tool_calls:
+        #     print_formatted_text(
+        #         FormattedText([(
+        #             "class:dimmed",
+        #             "\n\n[planner] calling:"
+        #         )]),
+        #         style=_dim_style,
+        #         end="\n"
+        #     )
+        #     for tc in response.tool_calls:
+        #         print_formatted_text(
+        #             FormattedText([(
+        #                 "class:dimmed", 
+        #                 f">> {str(tc['name'])}({str(tc['args'])})"
+        #             )]),
+        #             style=_dim_style,
+        #             end="\n"
+        #         )
                 # print("Tool: ", tc['name'])
                 # print("Arguments: ", tc['args'])
         
@@ -103,39 +103,40 @@ def create_agent_node(model: Runnable):
                 message_to_send += [plan_message] + [plan_instruction]
                 
                 
-                
-        # if not state.get("todo"):
-        #     message_to_send.append(SystemMessage(content=TODO_INSTRUCTIONS))
-            
-        # if state.get("todo"):
-        #     turns_since_update = _turns_since_todo_update(state["coder_messages"])
-        #     if turns_since_update >= 10:
-        #         message_to_send.append(
-        #             SystemMessage(content=_format_todo_reminder(state["todo"]))
-        #         )
+        # print(f"\n\n\n------ CONTEXT START------\n\n")
+        # for mess in message_to_send:
+        #     if isinstance(mess, SystemMessage):
+        #         print("SystemMessage: ", end="")
+        #     if isinstance(mess, HumanMessage):
+        #         print("HumanMessage: ", end="")
+        #     if isinstance(mess, ToolMessage):
+        #         print("ToolMessage: ", end="")
+        #     print(mess)
+        #     print()
+        # print("\n\n------ CONTEXT END ------\n\n\n")
         
         response: AIMessage = await model.with_config(config).ainvoke(message_to_send)
         
         # print("\n\nCODER------AI-RESPONSE:\n.\n.\n", response, "\n\n")
         
-        if response.tool_calls:
-            print_formatted_text(
-                FormattedText([(
-                    "class:dimmed",
-                    "\n\n[coder] calling:"
-                )]),
-                style=_dim_style,
-                end="\n"
-            )
-            for tc in response.tool_calls:
-                print_formatted_text(
-                    FormattedText([(
-                        "class:dimmed", 
-                        f">> {str(tc['name'])}({str(tc['args'])})"
-                    )]),
-                    style=_dim_style,
-                    end="\n"
-                )
+        # if response.tool_calls:
+        #     print_formatted_text(
+        #         FormattedText([(
+        #             "class:dimmed",
+        #             "\n\n[coder] calling:"
+        #         )]),
+        #         style=_dim_style,
+        #         end="\n"
+        #     )
+        #     for tc in response.tool_calls:
+        #         print_formatted_text(
+        #             FormattedText([(
+        #                 "class:dimmed", 
+        #                 f">> {str(tc['name'])}({str(tc['args'])})"
+        #             )]),
+        #             style=_dim_style,
+        #             end="\n"
+        #         )
         
         return {
             "coder_messages": [response],
@@ -341,11 +342,6 @@ def create_tool_execute_node(tool_registry: ToolRegistry):
         }
         if todo_update is not None:
             update["todo"] = todo_update
-            
-        # if interrupted:
-        #     # Save the state update, then END the graph invocation.
-        #     # Control returns to main(); the prompt shows again.
-        #     return Command(update=update, goto=END)
         
         return update