Connects your MCP client to the AI Incident Database, a public repository tracking real-world AI safety incidents and failures. You get three tools: fetch full incident records with linked reports, list recently added incidents, and browse the taxonomies used to classify them (CSET-AIID, GMF, RAIC). Useful when you're researching AI safety case studies, looking for precedents in automated system failures, or building risk assessments that need concrete examples. Hosted via Pipeworx's gateway, which offers both a plain English query interface through ask_pipeworx and the option to call tools directly. Free tier available, runs over streamable HTTP.
AI Incident Database (AIID) MCP
Part of Pipeworx — an MCP gateway connecting AI agents to 673+ live data sources.
| Tool | Description |
|---|---|
get_incident | Fetch full incident record with linked reports. |
list_recent | Most recently added incidents. |
list_taxonomies | Taxonomies used to classify incidents (CSET-AIID, GMF, RAIC). |
Add to your MCP client (Claude Desktop, Cursor, Windsurf, etc.):
{
"mcpServers": {
"ai-incident-db": {
"url": "https://gateway.pipeworx.io/ai-incident-db/mcp"
}
}
}
Or connect to the full Pipeworx gateway for access to all 673+ data sources:
{
"mcpServers": {
"pipeworx": {
"url": "https://gateway.pipeworx.io/mcp"
}
}
}
Instead of calling tools directly, you can ask questions in plain English:
ask_pipeworx({ question: "your question about Ai Incident Db data" })
The gateway picks the right tool and fills the arguments automatically.
MIT
hovecapital/read-only-local-postgres-mcp-server
cocaxcode/database-mcp
io.github.infoinlet-marketplace/mcp-mysql
io.github.cybeleri/database-admin
io.github.yash-0620/postgres-mcp-secured