prompt-copilot-cli 0.1.1__tar.gz

prompt_copilot_cli-0.1.1/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 Jason Li
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

prompt_copilot_cli-0.1.1/MANIFEST.in

@@ -0,0 +1,4 @@
+include README.md
+include README.zh-CN.md
+include LICENSE
+include requirements.txt

prompt_copilot_cli-0.1.1/PKG-INFO

@@ -0,0 +1,222 @@
+Metadata-Version: 2.4
+Name: prompt-copilot-cli
+Version: 0.1.1
+Summary: A lightweight terminal-based coding agent with file tools, command execution, multimodal image support, and MCP integration.
+Author: Jason Li
+License-Expression: MIT
+Project-URL: Homepage, https://github.com/woshiliyihang/prompt-copilot-cli
+Project-URL: Repository, https://github.com/woshiliyihang/prompt-copilot-cli
+Project-URL: Issues, https://github.com/woshiliyihang/prompt-copilot-cli/issues
+Keywords: cli,agent,openai,mcp,coding-assistant
+Classifier: Development Status :: 4 - Beta
+Classifier: Intended Audience :: Developers
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Topic :: Software Development :: Libraries
+Classifier: Topic :: Utilities
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: mcp==1.28.0
+Requires-Dist: openai==1.109.1
+Requires-Dist: prompt_toolkit==3.0.52
+Requires-Dist: rich==15.0.0
+Dynamic: license-file
+
+# Prompt Copilot CLI
+
+Prompt Copilot CLI is a lightweight terminal-based coding agent for local development workflows. It combines an OpenAI-compatible model with a set of practical tools for file operations, shell commands, Python execution, multimodal image handling, and MCP integrations.
+
+It is designed for developers who want an interactive coding assistant that can inspect a workspace, edit files, run commands, and help turn multi-step conversations into a final, actionable prompt.
+
+## ✨ Features
+
+- Interactive CLI experience in the terminal
+- Persistent session history and conversation logs
+- File-system tools for reading, writing, deleting, renaming, copying, and recursive directory listing
+- Shell command execution and Python script execution
+- Image support for vision-capable models via image-to-base64 conversion
+- MCP tool integration for extending the agent with external tools
+- Task workflow with `/task-start` and `/task-end` to generate a polished final prompt
+
+## 🚀 Quick Start
+
+### 1. Install dependencies
+
+```powershell
+py -m pip install -r requirements.txt
+```
+
+### 2. Configure the model
+
+On first launch, the project creates a configuration file at:
+
+- Windows: `%USERPROFILE%\.prompt-copilot\config.json`
+- Linux/macOS: `~/.prompt-copilot/config.json`
+
+Example:
+
+```json
+{
+  "model": "gpt-4o-mini",
+  "base_url": "http://127.0.0.1:11434/v1",
+  "api_key": "dummy",
+  "temperature": 0.2,
+  "debug": false,
+  "mcp": {
+    "enabled": true,
+    "servers": []
+  }
+}
+```
+
+### 3. MCP configuration (optional)
+
+The agent can discover and use external MCP tools through the `mcp.servers` array. This is useful when you want to extend the agent with tools such as web search, filesystem helpers, or other local services.
+
+Example configuration:
+
+```json
+{
+  "mcp": {
+    "enabled": true,
+    "servers": [
+      {
+        "name": "bing",
+        "command": "npx",
+        "args": ["-y", "bing-cn-mcp"]
+      },
+      {
+        "name": "open-websearch-http",
+        "transport": "http",
+        "url": "http://127.0.0.1:3000/mcp"
+      }
+    ]
+  }
+}
+```
+
+How it works:
+
+- The first server uses a local stdio-based MCP server launched by `npx`.
+- The second server connects to an HTTP MCP endpoint at the given URL.
+- Once discovered, the tools exposed by these servers become callable by the agent during a session.
+- If `enabled` is set to `false` or the server list is empty, no MCP tools will be loaded.
+
+### 4. Run the agent
+
+Interactive mode:
+
+```powershell
+py main.py
+```
+
+One-off task mode:
+
+```powershell
+py main.py -t "Create a simple HTML landing page" -d ./workspace -l en
+```
+
+## 🧭 Usage Guide
+
+### Interactive commands
+
+Once the CLI starts, you can use these commands:
+
+- `/exit` — quit the program
+- `/clear` — clear local session history
+- `/task-start` — start a task context for later summarization
+- `/task-end` — generate a final optimized prompt and save it to `last-prompt.md`
+
+### Common startup options
+
+```powershell
+py main.py -h
+```
+
+Key options:
+
+- `-t, --task` — one-off task content
+- `-d, --workdir` — working directory
+- `-l, --lang` — language (`zh` or `en`)
+- `-amc, --agent-messages-count` — number of messages kept in agent history
+- `-rd, --request-delay` — delay between model requests in seconds
+- `-hc, --history-count` — number of rounds kept in conversation history
+- `--reset-session` — reset persisted session history
+
+### Example workflows
+
+#### 1. Ask the agent to inspect a project
+
+```powershell
+py main.py -t "Inspect this repository and summarize the main structure" -d ./workspace
+```
+
+#### 2. Ask the agent to edit files and run tests
+
+```powershell
+py main.py -t "Update the code, then run the relevant test suite" -d ./workspace
+```
+
+#### 3. Ask the agent to analyze an image
+
+If your model supports vision, the agent can use the built-in image tool to read an image file and convert it to base64 for multimodal input.
+
+Example prompt:
+
+```text
+Please inspect the image in ./workspace/demo.png and tell me what numbers or text are visible.
+```
+
+## 🛠 Tool capabilities
+
+The agent can call the following tools:
+
+- File tools
+  - `read_file`
+  - `write_file`
+  - `delete_file`
+  - `create_directory`
+  - `delete_directory`
+  - `rename_path`
+  - `copy_file`
+  - `list_dir` (with recursive option)
+- Execution tools
+  - `execute_command`
+  - `execute_python_script`
+- Multimodal tools
+  - `read_image_as_base64`
+
+## 🧠 Task flow
+
+The project supports a lightweight task-iteration workflow:
+
+1. Start a round with `/task-start`
+2. Continue interacting with the agent to clarify requirements or refine the task
+3. Finish with `/task-end`
+4. The agent writes the final prompt to `last-prompt.md`
+
+This is useful when you want to turn a long back-and-forth conversation into a compact, executable prompt.
+
+## 📁 Project structure
+
+```text
+.
+├── main.py
+├── requirements.txt
+├── README.md
+├── README.zh-CN.md
+├── tests/
+└── workspace/
+```
+
+## 🤝 Contributing
+
+Contributions are welcome. Please feel free to open an issue or submit a pull request if you have suggestions, bug reports, or new workflow ideas.
+
+## 📄 License
+
+This project does not currently declare a specific license. If you plan to distribute or reuse it publicly, please add an appropriate license file.

prompt_copilot_cli-0.1.1/README.md

@@ -0,0 +1,194 @@
+# Prompt Copilot CLI
+
+Prompt Copilot CLI is a lightweight terminal-based coding agent for local development workflows. It combines an OpenAI-compatible model with a set of practical tools for file operations, shell commands, Python execution, multimodal image handling, and MCP integrations.
+
+It is designed for developers who want an interactive coding assistant that can inspect a workspace, edit files, run commands, and help turn multi-step conversations into a final, actionable prompt.
+
+## ✨ Features
+
+- Interactive CLI experience in the terminal
+- Persistent session history and conversation logs
+- File-system tools for reading, writing, deleting, renaming, copying, and recursive directory listing
+- Shell command execution and Python script execution
+- Image support for vision-capable models via image-to-base64 conversion
+- MCP tool integration for extending the agent with external tools
+- Task workflow with `/task-start` and `/task-end` to generate a polished final prompt
+
+## 🚀 Quick Start
+
+### 1. Install dependencies
+
+```powershell
+py -m pip install -r requirements.txt
+```
+
+### 2. Configure the model
+
+On first launch, the project creates a configuration file at:
+
+- Windows: `%USERPROFILE%\.prompt-copilot\config.json`
+- Linux/macOS: `~/.prompt-copilot/config.json`
+
+Example:
+
+```json
+{
+  "model": "gpt-4o-mini",
+  "base_url": "http://127.0.0.1:11434/v1",
+  "api_key": "dummy",
+  "temperature": 0.2,
+  "debug": false,
+  "mcp": {
+    "enabled": true,
+    "servers": []
+  }
+}
+```
+
+### 3. MCP configuration (optional)
+
+The agent can discover and use external MCP tools through the `mcp.servers` array. This is useful when you want to extend the agent with tools such as web search, filesystem helpers, or other local services.
+
+Example configuration:
+
+```json
+{
+  "mcp": {
+    "enabled": true,
+    "servers": [
+      {
+        "name": "bing",
+        "command": "npx",
+        "args": ["-y", "bing-cn-mcp"]
+      },
+      {
+        "name": "open-websearch-http",
+        "transport": "http",
+        "url": "http://127.0.0.1:3000/mcp"
+      }
+    ]
+  }
+}
+```
+
+How it works:
+
+- The first server uses a local stdio-based MCP server launched by `npx`.
+- The second server connects to an HTTP MCP endpoint at the given URL.
+- Once discovered, the tools exposed by these servers become callable by the agent during a session.
+- If `enabled` is set to `false` or the server list is empty, no MCP tools will be loaded.
+
+### 4. Run the agent
+
+Interactive mode:
+
+```powershell
+py main.py
+```
+
+One-off task mode:
+
+```powershell
+py main.py -t "Create a simple HTML landing page" -d ./workspace -l en
+```
+
+## 🧭 Usage Guide
+
+### Interactive commands
+
+Once the CLI starts, you can use these commands:
+
+- `/exit` — quit the program
+- `/clear` — clear local session history
+- `/task-start` — start a task context for later summarization
+- `/task-end` — generate a final optimized prompt and save it to `last-prompt.md`
+
+### Common startup options
+
+```powershell
+py main.py -h
+```
+
+Key options:
+
+- `-t, --task` — one-off task content
+- `-d, --workdir` — working directory
+- `-l, --lang` — language (`zh` or `en`)
+- `-amc, --agent-messages-count` — number of messages kept in agent history
+- `-rd, --request-delay` — delay between model requests in seconds
+- `-hc, --history-count` — number of rounds kept in conversation history
+- `--reset-session` — reset persisted session history
+
+### Example workflows
+
+#### 1. Ask the agent to inspect a project
+
+```powershell
+py main.py -t "Inspect this repository and summarize the main structure" -d ./workspace
+```
+
+#### 2. Ask the agent to edit files and run tests
+
+```powershell
+py main.py -t "Update the code, then run the relevant test suite" -d ./workspace
+```
+
+#### 3. Ask the agent to analyze an image
+
+If your model supports vision, the agent can use the built-in image tool to read an image file and convert it to base64 for multimodal input.
+
+Example prompt:
+
+```text
+Please inspect the image in ./workspace/demo.png and tell me what numbers or text are visible.
+```
+
+## 🛠 Tool capabilities
+
+The agent can call the following tools:
+
+- File tools
+  - `read_file`
+  - `write_file`
+  - `delete_file`
+  - `create_directory`
+  - `delete_directory`
+  - `rename_path`
+  - `copy_file`
+  - `list_dir` (with recursive option)
+- Execution tools
+  - `execute_command`
+  - `execute_python_script`
+- Multimodal tools
+  - `read_image_as_base64`
+
+## 🧠 Task flow
+
+The project supports a lightweight task-iteration workflow:
+
+1. Start a round with `/task-start`
+2. Continue interacting with the agent to clarify requirements or refine the task
+3. Finish with `/task-end`
+4. The agent writes the final prompt to `last-prompt.md`
+
+This is useful when you want to turn a long back-and-forth conversation into a compact, executable prompt.
+
+## 📁 Project structure
+
+```text
+.
+├── main.py
+├── requirements.txt
+├── README.md
+├── README.zh-CN.md
+├── tests/
+└── workspace/
+```
+
+## 🤝 Contributing
+
+Contributions are welcome. Please feel free to open an issue or submit a pull request if you have suggestions, bug reports, or new workflow ideas.
+
+## 📄 License
+
+This project does not currently declare a specific license. If you plan to distribute or reuse it publicly, please add an appropriate license file.

prompt_copilot_cli-0.1.1/README.zh-CN.md

@@ -0,0 +1,194 @@
+# Prompt Copilot CLI
+
+Prompt Copilot CLI 是一个轻量级的终端编程 Agent，适合本地开发场景。它将 OpenAI 兼容模型与一组实用工具结合起来，包括文件操作、命令执行、Python 脚本运行、多模态图片处理以及 MCP 工具扩展。
+
+它适合希望在终端中进行交互式开发、查看项目结构、修改文件、运行命令，并把多轮对话整理成最终可执行提示词的开发者。
+
+## ✨ 功能特性
+
+- 终端交互式 CLI
+- 持久化会话历史与对话记录
+- 文件系统工具：读取、写入、删除、重命名、复制、递归列出目录
+- 命令执行与 Python 脚本执行
+- 支持视觉模型的图片处理：将图片读取并转成 base64
+- MCP 工具集成，便于扩展外部能力
+- 支持 `/task-start` 和 `/task-end` 的任务流，生成最终优化提示词
+
+## 🚀 快速开始
+
+### 1. 安装依赖
+
+```powershell
+py -m pip install -r requirements.txt
+```
+
+### 2. 配置模型
+
+首次运行时，项目会在如下位置生成配置文件：
+
+- Windows: `%USERPROFILE%\.prompt-copilot\config.json`
+- Linux/macOS: `~/.prompt-copilot/config.json`
+
+示例配置：
+
+```json
+{
+  "model": "gpt-4o-mini",
+  "base_url": "http://127.0.0.1:11434/v1",
+  "api_key": "dummy",
+  "temperature": 0.2,
+  "debug": false,
+  "mcp": {
+    "enabled": true,
+    "servers": []
+  }
+}
+```
+
+### 3. MCP 配置（可选）
+
+Agent 可以通过 `mcp.servers` 数组发现并调用外部 MCP 工具。这个能力适合接入网页搜索、外部服务、辅助工具等扩展能力。
+
+配置示例：
+
+```json
+{
+  "mcp": {
+    "enabled": true,
+    "servers": [
+      {
+        "name": "bing",
+        "command": "npx",
+        "args": ["-y", "bing-cn-mcp"]
+      },
+      {
+        "name": "open-websearch-http",
+        "transport": "http",
+        "url": "http://127.0.0.1:3000/mcp"
+      }
+    ]
+  }
+}
+```
+
+说明：
+
+- 第一个示例是通过 `npx` 启动一个本地 stdio 类型的 MCP 服务。
+- 第二个示例是连接到一个 HTTP 类型的 MCP 服务端点。
+- 一旦这些服务被发现，Agent 就可以在会话中调用它们暴露出来的工具。
+- 如果 `enabled` 设为 `false`，或者 `servers` 为空，则不会加载任何 MCP 工具。
+
+### 4. 启动 Agent
+
+交互模式：
+
+```powershell
+py main.py
+```
+
+单次任务模式：
+
+```powershell
+py main.py -t "创建一个简单的 HTML 落地页" -d ./workspace -l zh
+```
+
+## 🧭 使用说明
+
+### 交互命令
+
+启动后可以使用以下命令：
+
+- `/exit`：退出程序
+- `/clear`：清空本地会话历史
+- `/task-start`：开始任务上下文
+- `/task-end`：生成最终优化提示词并写入 `last-prompt.md`
+
+### 常用启动参数
+
+```powershell
+py main.py -h
+```
+
+主要参数：
+
+- `-t, --task`：一次性任务内容
+- `-d, --workdir`：工作目录
+- `-l, --lang`：语言（`zh` 或 `en`）
+- `-amc, --agent-messages-count`：保留在 Agent 历史中的消息数量
+- `-rd, --request-delay`：模型请求之间的间隔秒数
+- `-hc, --history-count`：会话历史保留的轮次数量
+- `--reset-session`：重置本地持久化会话记录
+
+### 示例场景
+
+#### 1. 检查项目结构
+
+```powershell
+py main.py -t "检查这个仓库并总结主要结构" -d ./workspace
+```
+
+#### 2. 编辑文件并执行测试
+
+```powershell
+py main.py -t "修改代码并运行相关测试" -d ./workspace
+```
+
+#### 3. 分析图片内容
+
+如果你的模型支持视觉能力，Agent 可以调用图片工具读取图片并将其转换成 base64 供多模态模型分析。
+
+例如：
+
+```text
+请看一下 ./workspace/demo.png 中能看到哪些数字或文字。
+```
+
+## 🛠 工具能力
+
+Agent 可以调用这些工具：
+
+- 文件工具
+  - `read_file`
+  - `write_file`
+  - `delete_file`
+  - `create_directory`
+  - `delete_directory`
+  - `rename_path`
+  - `copy_file`
+  - `list_dir`（支持递归）
+- 执行工具
+  - `execute_command`
+  - `execute_python_script`
+- 多模态工具
+  - `read_image_as_base64`
+
+## 🧠 任务流
+
+项目支持一种轻量的任务迭代流程：
+
+1. 用 `/task-start` 开始一轮任务
+2. 与 Agent 连续沟通，澄清需求或修正目标
+3. 用 `/task-end` 结束任务
+4. 最终提示词被写入 `last-prompt.md`
+
+这对于将多轮对话整理成一条可直接执行的提示词非常有帮助。
+
+## 📁 项目结构
+
+```text
+.
+├── main.py
+├── requirements.txt
+├── README.md
+├── README.zh-CN.md
+├── tests/
+└── workspace/
+```
+
+## 🤝 参与贡献
+
+欢迎提交 Issue、PR 或建议。如果你有新的功能想法、错误反馈或改进建议，欢迎一起参与。
+
+## 📄 许可证
+
+当前项目尚未声明特定许可证。如果你准备公开分发或复用，请补充合适的许可证文件。