Lore Context
Give your agent governed long-term memory, evidence traces, eval hooks, and portable context by proxying to a Lore API from MCP.
Overview
Lore Context MCP is a MCP server for the Build phase that proxies governed agent memory, evidence ledger traces, evals, and portable context to the Lore API.
What is this MCP server?
- Governed AI-agent memory with API-backed persistence
- Evidence Ledger traces for auditable agent decisions
- Eval-oriented tooling for measuring agent behavior
- Portable context tools across sessions and projects
- Configurable via LORE_API_URL (required) and optional LORE_API_KEY
- Version 0.6.0-alpha.1
- npm identifier @lore-context/server
- Three documented env vars: LORE_API_URL, LORE_API_KEY, LORE_MCP_TRANSPORT (default sdk)
Community signal: 4 GitHub stars.
What problem does it solve?
Agents lose context across sessions and lack auditable memory unless you bolt on a custom store and trace pipeline.
Who is it for?
Builders running a Lore API locally or hosted who want MCP-native memory and traces in Claude Code or Cursor.
Skip if: Anyone who wants zero external service, no API to run, or fully stable GA-only dependencies.
What do I get? / Deliverables
After install, tool calls hit your Lore API so agents can reuse governed memory and evidence-backed context instead of one-off chat history.
- MCP bridge from agent to Lore memory and trace tools
- Session-spanning portable context without custom MCP authoring
Recommended MCP Servers
Journey fit
Agent memory and context governance matter most while building agent features, but the same tools support iteration and review later. Agent-tooling is the canonical shelf for MCP servers that extend what coding agents remember and audit across sessions.
How it compares
Governed memory MCP proxy, not a simple vector-ingest skill or static RAG snippet pack.
Common Questions / FAQ
Who is Lore Context MCP for?
Solo builders and small teams building agent workflows who already use or can run the Lore API and want MCP-accessible memory and traces.
When should I use Lore Context MCP?
Use it while building agent features and when iterating on prompts that need durable, governed context and eval-friendly evidence.
How do I add Lore Context MCP to my agent?
Set LORE_API_URL (required), optionally LORE_API_KEY, add @lore-context/server via npx stdio in your MCP config, and ensure the Lore API is reachable.