data-ai MCP Server
An autonomous AI agent that reads your whole project, runs commands, and writes & verifies code on its own — going deeper than autocomplete. Bring any model, hosted or local.
Discovered via agent-topic:ai-agent and last synced 3mo ago.
1. Install the package
npx @modelcontextprotocol/server-filesystem /path
2. Add to claude_desktop_config.json
{
"mcpServers": {
"codeep": {
"command": "npx",
"args": [
"codeep"
]
}
}
}Config file location: ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) / %APPDATA%\Claude\claude_desktop_config.json (Windows)
Search for patterns in code
Build
Cancel current request / Stop agent
German (Deutsch)
DeepSeek API key
Edit specific text in files
Before every agent tool call (including MCP `<server>__<tool>`)
Delete files or directories
After `write_file` / `edit_file` succeeds (built-in only)
List directory contents
Before the `/commit` skill stages anything
fallbacks
Fetch content from URLs
Binary
Language
Japanese (日本語)
Google AI (Gemini) API key
`Never`, `Dangerous` (default), or `Always`
Create folders
When any tool call fails (built-in or MCP)
Behavior
Arabic (العربية)
Z.AI China API key
Request timeout
Description
info
```bash # Example for macOS Apple Silicon: curl -fsSL https://github.com/VladoIvankovic/Codeep/releases/latest/download/codeep-macos-arm64 -o codeep chmod +x codeep sudo mv codeep /usr/local/bin/ ``` ## Quick Start ```bash # Navigate to your project directory cd /path/to/your/project # Start Codeep codeep # On first run, enter your API key # Get one at: https://z.ai/subscribe?ic=NXYNXZOV14 ``` After installation, `codeep` is available globally in your terminal. Simply run it from any project directory to start coding with AI assistance. ## Commands ### General
Auto-detect (matches user's language)
Read file contents
Information
Description
English
Croatian (Hrvatski)
MiniMax China API key
Run shell commands
Execute without asking
Create or overwrite files
When
French (Français)
Russian (Русский)
MiniMax (international) API key
Maximum agent iterations (10-500). Modern models finish typical tasks in 3–8 iterations — raise for large autonomous refactors.
Ask before every action
Hindi (हिन्दी)
Description
Default
Action
Z.AI (international) API key
0.7
Timeout per agent API call (auto-adjusted for complexity)
Chinese (中文)
Anthropic Claude API key
`ON` = agent runs automatically (requires write permission via `/grant`), `Manual` = use /agent
OpenAI API key
Spanish (Español)
Location
`Off`, `Build only`, `Typecheck only`, `Test only`, or `Build + Typecheck + Test`
Submit message
Portuguese (Português)
Maximum response length
Max attempts to auto-fix errors when Auto-Verify is enabled. More than 1 usually means the agent is stuck — bail and let the user decide.