Skip to content
Back to search

just-prompt

just-prompt is an MCP server that provides a unified interface to top LLM providers (OpenAI, Anthropic, Google Gemini, Groq, DeepSeek, and Ollama)

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/disler--just-prompt/badge.svg)](/listings/disler--just-prompt)

HTML

<a href="/listings/disler--just-prompt"><img src="/api/listings/disler--just-prompt/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

# Clone the repository
git clone https://github.com/yourusername/just-prompt.git
cd just-prompt

# Install with pip
uv sync

Environment Variables

Create a `.env` file with your API keys (you can copy the `.env.sample` file):

cp .env.sample .env

Then edit the `.env` file to add your API keys (or export them in your shell):

OPENAI_API_KEY=your_openai_api_key_here
ANTHROPIC_API_KEY=your_anthropic_api_key_here
GEMINI_API_KEY=your_gemini_api_key_here
GROQ_API_KEY=your_groq_api_key_here
DEEPSEEK_API_KEY=your_deepseek_api_key_here
OLLAMA_HOST=http://localhost:11434

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 330 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. # Just Prompt - A lightweight MCP server for LLM providers
  2. `just-prompt` is a Model Control Protocol (MCP) server that provides a unified interface to various Large Language Model (LLM) providers including OpenAI, Anthropic, Google Gemini, Groq, DeepSeek, and Ollama. See how we use the `ceo_and_board` tool to make [hard decisions easy with o3 here](https://youtu.be/LEMLntjfihA).
  3. <img src="images/just-prompt-logo.png" alt="Just Prompt Logo" width="700" height="auto">
  4. <img src="images/o3-as-a-ceo.png" alt="Just Prompt Logo" width="700" height="auto">
  5. ## Tools
  6. The following MCP tools are available in the server:
  7. - **`prompt`**: Send a prompt to multiple LLM models
  8. - Parameters:
  9. - `text`: The prompt text
  10. - `models_prefixed_by_provider` (optional): List of models with provider prefixes. If not provided, uses default models.
  11. - **`prompt_from_file`**: Send a prompt from a file to multiple LLM models
  12. - Parameters:
  13. - `abs_file_path`: Absolute path to the file containing the prompt (must be an absolute path, not relative)
  14. - `models_prefixed_by_provider` (optional): List of models with provider prefixes. If not provided, uses default models.
  15. - **`prompt_from_file_to_file`**: Send a prompt from a file to multiple LLM models and save responses as markdown files
  16. - Parameters:
  17. - `abs_file_path`: Absolute path to the file containing the prompt (must be an absolute path, not relative)
  18. - `models_prefixed_by_provider` (optional): List of models with provider prefixes. If not provided, uses default models.
  19. - `abs_output_dir` (default: "."): Absolute directory path to save the response markdown files to (must be an absolute path, not relative)
  20. - **`ceo_and_board`**: Send a prompt to multiple 'board member' models and have a 'CEO' model make a decision based on their responses
  21. - Parameters:
  22. - `abs_file_path`: Absolute path to the file containing the prompt (must be an absolute path, not relative)
  23. - `models_prefixed_by_provider` (optional): List of models with provider prefixes to act as board members. If not provided, uses default models.
  24. - `abs_output_dir` (default: "."): Absolute directory path to save the response files and CEO decision (must be an absolute path, not relative)
  25. - `ceo_model` (default: "openai:o3"): Model to use for the CEO decision in format "provider:model"
  26. - **`list_providers`**: List all available LLM providers
  27. - Parameters: None
  28. - **`list_models`**: List all available models for a specific LLM provider
  29. - Parameters:
  30. - `provider`: Provider to list models for (e.g., 'openai' or 'o')
  31. ## Provider Prefixes
  32. > every model must be prefixed with the provider name
  33. >
  34. > use the short name for faster referencing
  35. - `o` or `openai`: OpenAI
  36. - `o:gpt-4o-mini`
  37. - `openai:gpt-4o-mini`
  38. - `a` or `anthropic`: Anthropic
  39. - `a:claude-3-5-haiku`
  40. - `anthropic:claude-3-5-haiku`
  41. - `g` or `gemini`: Google Gemini
  42. - `g:gemini-2.5-pro-exp-03-25`
  43. - `gemini:gemini-2.5-pro-exp-03-25`
  44. - `q` or `groq`: Groq
  45. - `q:llama-3.1-70b-versatile`
  46. - `groq:llama-3.1-70b-versatile`
  47. - `d` or `deepseek`: DeepSeek
  48. - `d:deepseek-coder`
  49. - `deepseek:deepseek-coder`
  50. - `l` or `ollama`: Ollama
  51. - `l:llama3.1`
  52. - `ollama:llama3.1`
  53. ## Features
  54. - Unified API for multiple LLM providers
  55. - Support for text prompts from strings or files
  56. - Run multiple models in parallel
  57. - Automatic model name correction using the first model in the `--default-models` list
  58. - Ability to save responses to files
  59. - Easy listing of available providers and models
  60. ## Installation
  61. ```bash
  62. # Clone the repository
  63. git clone https://github.com/yourusername/just-prompt.git
  64. cd just-prompt
  65. # Install with pip
  66. uv sync
  67. ```
  68. ### Environment Variables
  69. Create a `.env` file with your API keys (you can copy the `.env.sample` file):
  70. ```bash
  71. cp .env.sample .env
  72. ```
  73. Then edit the `.env` file to add your API keys (or export them in your shell):
  74. ```
  75. OPENAI_API_KEY=your_openai_api_key_here
  76. ANTHROPIC_API_KEY=your_anthropic_api_key_here
  77. GEMINI_API_KEY=your_gemini_api_key_here
  78. GROQ_API_KEY=your_groq_api_key_here
  79. DEEPSEEK_API_KEY=your_deepseek_api_key_here
  80. OLLAMA_HOST=http://localhost:11434
  81. ```
  82. ## Claude Code Installation
  83. > In all these examples, replace the directory with the path to the just-prompt directory.
  84. Default models set to `openai:o3:high`, `openai:o4-mini:high`, `anthropic:claude-opus-4-20250514`, `anthropic:claude-sonnet-4-20250514`, `gemini:gemini-2.5-pro-preview-03-25`, and `gemini:gemini-2.5-flash-preview-04-17`.
  85. If you use Claude Code right out of the repository you can see in the .mcp.json file we set the default models to...
  86. ```
  87. {
  88. "mcpServers": {
  89. "just-prompt": {
  90. "type": "stdio",
  91. "command": "uv",
  92. "args": [
  93. "--directory",
  94. ".",
  95. "run",
  96. "just-prompt",
  97. "--default-models",
  98. "openai:o3:high,openai:o4-mini:high,anthropic:claude-opus-4-20250514,anthropic:claude-sonnet-4-20250514,gemini:gemini-2.5-pro-preview-03-25,gemini:gemini-2.5-flash-preview-04-17"
  99. ],
  100. "env": {}
  101. }
  102. }
  103. }
  104. ```
  105. The `--default-models` parameter sets the models to use when none are explicitly provided to the API endpoints. The first model in the list is also used for model name correction when needed. This can be a list of models separated by commas.
  106. When starting the server, it will automatically check which API keys are available in your environment and inform you which providers you can use. If a key is missing, the provider will be listed as unavailable, but the server will still start and can be used with the providers that are available.
  107. ### Using `mcp add-json`
  108. Copy this and paste it into claude code with BUT don't run until you copy the json
  109. ```
  110. claude mcp add just-prompt "$(pbpaste)"
  111. ```
  112. JSON to copy
  113. ```
  114. {
  115. "command": "uv",
  116. "args": ["--directory", ".", "run", "just-prompt"]
  117. }
  118. ```
  119. With a custom default model set to `openai:gpt-4o`.
  120. ```
  121. {
  122. "command": "uv",
  123. "args": ["--directory", ".", "run", "just-prompt", "--default-models", "openai:gpt-4o"]
  124. }
  125. ```
  126. With multiple default models:
  127. ```
  128. {
  129. "command": "uv",
  130. "args": ["--directory", ".", "run", "just-prompt", "--default-models", "openai:o3:high,openai:o4-mini:high,anthropic:claude-opus-4-20250514,anthropic:claude-sonnet-4-20250514,gemini:gemini-2.5-pro-preview-03-25,gemini:gemini-2.5-flash-preview-04-17"]
  131. }
  132. ```
  133. ### Using `mcp add` with project scope
  134. ```bash
  135. # With default models
  136. claude mcp add just-prompt -s project \
  137. -- \
  138. uv --directory . \
  139. run just-prompt
  140. # With custom default model
  141. claude mcp add just-prompt -s project \
  142. -- \
  143. uv --directory . \
  144. run just-prompt --default-models "openai:gpt-4o"
  145. # With multiple default models
  146. claude mcp add just-prompt -s user \
  147. -- \
  148. uv --directory . \
  149. run just-prompt --default-models "openai:o3:high,openai:o4-mini:high,anthropic:claude-opus-4-20250514,anthropic:claude-sonnet-4-20250514,gemini:gemini-2.5-pro-preview-03-25,gemini:gemini-2.5-flash-preview-04-17"
  150. ```
  151. ## `mcp remove`
  152. claude mcp remove just-prompt
  153. ## Running Tests
  154. ```bash
  155. uv run pytest
  156. ```
  157. ## Codebase Structure
  158. ```
  159. .
  160. ├── ai_docs/ # Documentation for AI model details
  161. │ ├── extending_thinking_sonny.md
  162. │ ├── llm_providers_details.xml
  163. │ ├── openai-reasoning-effort.md
  164. │ └── pocket-pick-mcp-server-example.xml
  165. ├── example_outputs/ # Example outputs from different models
  166. ├── list_models.py # Script to list available LLM models
  167. ├── prompts/ # Example prompt files
  168. ├── pyproject.toml # Python project configuration
  169. ├── specs/ # Project specifications
  170. │ ├── init-just-prompt.md
  171. │ ├── new-tool-llm-as-a-ceo.md
  172. │ └── oai-reasoning-levels.md
  173. ├── src/ # Source code directory
  174. │ └── just_prompt/
  175. │ ├── __init__.py
  176. │ ├── __main__.py
  177. │ ├── atoms/ # Core components
  178. │ │ ├── llm_providers/ # Individual provider implementations
  179. │ │ │ ├── anthropic.py
  180. │ │ │ ├── deepseek.py
  181. │ │ │ ├── gemini.py
  182. │ │ │ ├── groq.py
  183. │ │ │ ├── ollama.py
  184. │ │ │ └── openai.py
  185. │ │ └── shared/ # Shared utilities and data types
  186. │ │ ├── data_types.py
  187. │ │ ├── model_router.py
  188. │ │ ├── utils.py
  189. │ │ └── validator.py
  190. │ ├── molecules/ # Higher-level functionality
  191. │ │ ├── ceo_and_board_prompt.py
  192. │ │ ├── list_models.py
  193. │ │ ├── list_providers.py
  194. │ │ ├── prompt.py
  195. │ │ ├── prompt_from_file.py
  196. │ │ └── prompt_from_file_to_file.py
  197. │ ├── server.py # MCP server implementation
  198. │ └── tests/ # Test directory
  199. │ ├── atoms/ # Tests for atoms
  200. │ │ ├── llm_providers/
  201. │ │ └── shared/
  202. │ └── molecules/ # Tests for molecules
  203. │ ├── test_ceo_and_board_prompt.py
  204. │ ├── test_list_models.py
  205. │ ├── test_list_providers.py
  206. │ ├── test_prompt.py
  207. │ ├── test_prompt_from_file.py
  208. │ └── test_prompt_from_file_to_file.py
  209. └── ultra_diff_review/ # Diff review outputs
  210. ```
  211. ## Context Priming
  212. READ README.md, pyproject.toml, then run git ls-files, and 'eza --git-ignore --tree' to understand the context of the project.
  213. # Reasoning Effort with OpenAI o‑Series
  214. For OpenAI o‑series reasoning models (`o4-mini`, `o3-mini`, `o3`) you can
  215. control how much *internal* reasoning the model performs before producing a
  216. visible answer.
  217. Append one of the following suffixes to the model name (after the *provider*
  218. prefix):
  219. * `:low`   – minimal internal reasoning (faster, cheaper)
  220. * `:medium` – balanced (default if omitted)
  221. * `:high`  – thorough reasoning (slower, more tokens)
  222. Examples:
  223. * `openai:o4-mini:low`
  224. * `o:o4-mini:high`
  225. When a reasoning suffix is present, **just‑prompt** automatically switches to
  226. the OpenAI *Responses* API (when available) and sets the corresponding
  227. `reasoning.effort` parameter. If the installed OpenAI SDK is older, it
  228. gracefully falls back to the Chat Completions endpoint and embeds an internal
  229. system instruction to approximate the requested effort level.
  230. # Thinking Tokens with Claude
  231. The Anthropic Claude models `claude-opus-4-20250514` and `claude-sonnet-4-20250514` support extended thinking capabilities using thinking tokens. This allows Claude to do more thorough thought processes before answering.
  232. You can enable thinking tokens by adding a suffix to the model name in this format:
  233. - `anthropic:claude-opus-4-20250514:1k` - Use 1024 thinking tokens for Opus 4
  234. - `anthropic:claude-sonnet-4-20250514:4k` - Use 4096 thinking tokens for Sonnet 4
  235. - `anthropic:claude-opus-4-20250514:8000` - Use 8000 thinking tokens for Opus 4
  236. Notes:
  237. - Thinking tokens are supported for `claude-opus-4-20250514`, `claude-sonnet-4-20250514`, and `claude-3-7-sonnet-20250219` models
  238. - Valid thinking token budgets range from 1024 to 16000
  239. - Values outside this range will be automatically adjusted to be within range
  240. - You can specify the budget with k notation (1k, 4k, etc.) or with exact numbers (1024, 4096, etc.)
  241. # Thinking Budget with Gemini
  242. The Google Gemini model `gemini-2.5-flash-preview-04-17` supports extended thinking capabilities using thinking budget. This allows Gemini to perform more thorough reasoning before providing a response.
  243. You can enable thinking budget by adding a suffix to the model name in this format:
  244. - `gemini:gemini-2.5-flash-preview-04-17:1k` - Use 1024 thinking budget
  245. - `gemini:gemini-2.5-flash-preview-04-17:4k` - Use 4096 thinking budget
  246. - `gemini:gemini-2.5-flash-preview-04-17:8000` - Use 8000 thinking budget
  247. Notes:
  248. - Thinking budget is only supported for the `gemini-2.5-flash-preview-04-17` model
  249. - Valid thinking budget range from 0 to 24576
  250. - Values outside this range will be automatically adjusted to be within range
  251. - You can specify the budget with k notation (1k, 4k, etc.) or with exact numbers (1024, 4096, etc.)
  252. ## Resources
  253. - https://docs.anthropic.com/en/api/models-list?q=list+models
  254. - https://github.com/googleapis/python-genai
  255. - https://platform.openai.com/docs/api-reference/models/list
  256. - https://api-docs.deepseek.com/api/list-models
  257. - https://github.com/ollama/ollama-python
  258. - https://github.com/openai/openai-python
  259. ## Master AI Coding
  260. Learn to code with AI with foundational [Principles of AI Coding](https://agenticengineer.com/principled-ai-coding?y=jprompt)
  261. Follow the [IndyDevDan youtube channel](https://www.youtube.com/@indydevdan) for more AI coding tips and tricks.

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.