A persistent memory layer for Claude that survives conversation resets. It exposes tools for locking context as immutable snapshots, hybrid semantic search across stored knowledge, and NLI-based conflict detection to catch contradictions in your project documentation. You also get built-in ticket management with state machines, document ingestion with AI chunking, and multi-project isolation. Authentication runs through OAuth2 on first connect. Reach for this when you're tired of re-explaining your codebase every session or need Claude to remember decisions, architecture constraints, and project context across weeks of work. Connects via streamable-http to a hosted endpoint at mcp.stompy.ai.
Persistent memory system that prevents context loss between AI sessions. Store, search, and recall project knowledge across conversations.
Search @mcp stompy in the extensions gallery, or add manually:
// .vscode/mcp.json
{
"servers": {
"stompy": {
"type": "http",
"url": "https://mcp.stompy.ai"
}
}
}
Add to your Claude Desktop config (~/Library/Application Support/Claude/claude_desktop_config.json):
{
"mcpServers": {
"stompy": {
"url": "https://mcp.stompy.ai"
}
}
}
claude mcp add stompy --transport streamable-http https://mcp.stompy.ai
| Tool | Description |
|---|---|
lock_context | Store content as immutable versioned snapshot |
recall_context | Retrieve locked context by topic |
context_search | Hybrid semantic + keyword search |
ticket | CRUD + lifecycle for project tickets |
ticket_board | View ticket board grouped by status |
detect_conflicts | Detect contradictions using NLI |
ingest_document | Ingest documents with AI processing |
project_create | Create a new project with isolated storage |
db_query | Read-only SQL queries against project database |
Stompy uses OAuth2 authentication. On first connection, your MCP client will prompt you to authorize via browser.
https://mcp.stompy.aiio.github.ericm1018/skillfm-llm-cost-optimizer-openai-anthropic-usage
io.github.mikerawsonnz/llm-orchestration-agent
io.github.mikerawsonnz/authenticated-llm-agent
labforgedev/copilot-memory-mcp
csoai-org/agent-prompt-injection-firewall-mcp
io.github.mikerawsonnz/authenticated-multi-llm-agent