Deadends Dev
@dbwls99706
About Deadends Dev
Structured failure knowledge infrastructure for AI agents — dead ends, workarounds, and error transition graphs.
Config
Add this server to your MCP-compatible client using the configuration below.
{
"mcpServers": {
"deadend": {
"command": "npx",
"args": [
"-y",
"@smithery/cli@latest",
"run",
"deadend/deadends-dev"
]
}
}
}Tools
No tools detected
Fetch the live tool list by running this server in a temporary sandbox using the button above.
Overview
What is Deadends Dev?
Deadends Dev is an MCP server that provides a database of known errors and dead ends for AI agents, covering both code failures and country-specific real-world rules. It helps agents avoid repeating known failures by offering deterministic, community-validated advice.
How to use Deadends Dev?
Install via pip with pip install deadends-dev, then invoke commands like deadends "error message". For MCP integration, add the server configuration to ~/.claude/claude_desktop_config.json or install via Smithery. The server provides 11 tools accessible via MCP clients.
Key features of Deadends Dev
- Deterministic answers with no hallucination
- Country-scoped data using ISO alpha-2 codes
- Primary-sourced from government and verifiable sources
- Community-validated fix success rates
- Sub-millisecond local regex matching
- Over 2,200 canon entries across 54 domains
Use cases of Deadends Dev
- Prevent agents from trying
sudo pip installthat fails 70% of the time - Avoid AI advice to tip in Japan where tipping is culturally refused
- Flag legal risks like Article 112 lèse-majesté in Thailand
- Handle error transitions where one fix triggers another error
- Stop agents from suggesting public displays of affection in UAE
FAQ from Deadends Dev
What makes Deadends Dev different from asking an LLM?
Deadends Dev provides deterministic answers (same query, same result) with no hallucination, country-scoped data, primary-sourced references, community-validated success rates, and sub-millisecond local response.
What are the practical limitations of Deadends Dev?
The dataset does not cover every error; missing cases can be added via issues, PRs, or the report_outcome tool. It prioritizes practical dead ends and workarounds over deep explanations. Reliability varies by domain and case; high-risk changes should be cross-verified with official sources.
How can I resolve "Unauthorized" errors when using the MCP server?
Common causes include wrong cwd path, running both local and remote (Smithery) modes simultaneously, or expired authentication. Ensure only one mode is active, verify the configuration file, and
Frequently asked questions
What makes Deadends Dev different from asking an LLM?
Deadends Dev provides deterministic answers (same query, same result) with no hallucination, country-scoped data, primary-sourced references, community-validated success rates, and sub-millisecond local response.
What are the practical limitations of Deadends Dev?
The dataset does not cover every error; missing cases can be added via issues, PRs, or the `report_outcome` tool. It prioritizes practical dead ends and workarounds over deep explanations. Reliability varies by domain and case; high-risk changes should be cross-verified with official sources.
How can I resolve "Unauthorized" errors when using the MCP server?
Common causes include wrong `cwd` path, running both local and remote (Smithery) modes simultaneously, or expired authentication. Ensure only one mode is active, verify the configuration file, and
Basic information
More Other MCP servers
🚀 Model Context Protocol (MCP) Curriculum for Beginners
microsoftThis open-source curriculum introduces the fundamentals of Model Context Protocol (MCP) through real-world, cross-language examples in .NET, Java, TypeScript, JavaScript, Rust and Python. Designed for developers, it focuses on practical techniques for building modular, scalable,

Glasswarp
GlasswarpSee and control a real Windows PC you own — from any MCP client, locally or remotely. Observe (UIA + screenshots), click/type/drag/scroll, launch apps, owner Live View. BYOH: your machine, your key.
Production-ready MCP integrations for AI applications
Klavis-AIKlavis AI: MCP integration platforms that let AI agents use tools reliably at any scale
Inbox Zero AI
elie222The world's best AI personal assistant for email. Open source app to help you reach inbox zero fast.
MaxKB
1Panel-dev🔥 MaxKB is an open-source platform for building enterprise-grade agents. 强大易用的开源企业级智能体平台。
Comments