Yantrikdb

YantrikDB — Cognitive memory for AI agents. Persistent semantic recall, knowledge graph, contradiction detection, and procedural learning. Ships as embeddable engine, network database, or MCP server.

Works with Claude Code, Cursor, Windsurf,Hermes Agent,Prime Agent, and any MCP-compatible client. Ships a portableAgent Skillsskill —skills/persistent-memory— that teaches any compliant harness the memory golden path.

Website:yantrikdb.com·Docs:yantrikdb.com/guides/mcp·GitHub:yantrikos/yantrikdb-mcp·Paper:Skill as Memory, Not Document

Every value on screen is the server's own answer over MCP — driver:docs/demo/demo.py, recorded withdocs/demo/demo.tape.

# Default — uses the engine's bundled 64-dim embedder. ~10 MB install, # ~80 ms cold start, no native ML deps. pip install yantrikdb-mcp # Optional: higher-quality 384-dim ONNX MiniLM-L6-v2 embedder (~150 MB install). # Auto-used when an existing pre-v0.6 database is detected. pip install 'yantrikdb-mcp[onnx]'

Upgrading from v0.5.x?Your existing database stays at 384 dim — install the[onnx]extra to keep using it transparently. New installs default to the lean bundled embedder. v0.7.0+ pins the engine migration fix automatically. SeeEmbedder backendsbelow.

{ "mcpServers": { "yantrikdb": { "command": "yantrikdb-mcp" } } }

Forward all tool calls to aYantrikDB HTTP clusterinstead of using an embedded engine. The MCP server is a thin stateless client — all memories live on the cluster, accessible from any machine.

{ "mcpServers": { "yantrikdb": { "command": "yantrikdb-mcp", "env": { "YANTRIKDB_SERVER_URL": "http://node1:7438,http://node2:7438", "YANTRIKDB_TOKEN": "ydb_your_database_token" } } } }

# Generate a secure API key export YANTRIKDB_API_KEY=$(python -c "import secrets; print(secrets.token_urlsafe(32))") # Start SSE server yantrikdb-mcp --transport sse --port 8420

{ "mcpServers": { "yantrikdb": { "type": "sse", "url": "http://your-server:8420/sse", "headers": { "Authorization": "Bearer YOUR_API_KEY" } } } }

RuntimeError: Existing DB has memories embedded with the 384-dim ONNX model, but ONNX deps are missing. Install with: pip install 'yantrikdb-mcp[onnx]'

For the multilingual backend, the engine downloadspotion-multilingual-128M(~460 MB tarball) fromgithub.com/yantrikos/yantrikdb-modelson first use. The download is SHA-256 verified, extracted into the engine's cache dir, and reused on subsequent starts. No extra Python deps required — the model runs entirely inside the Rust engine.

File-based memory (CLAUDE.md, memory files) loadseverythinginto context every conversation. YantrikDB recalls only what's relevant.

Selective recall is O(1). File-based memory is O(n).

- At 500 memories, file-based exceeds 32K context windows
- At 5,000, it doesn't fit inanycontext window — not even 200K
- YantrikDB stays at ~70 tokens per query, under 60ms latency
- Precisionimproveswith more data — the opposite of context stuffing

Run the benchmark yourself:python benchmarks/bench_token_savings.py

Recommended agent workflow (golden path)

The server injects a golden-path playbook into the agent's system prompt. Since v0.10.0 the default isdigest-first:
- Cold start — one call.session(action="digest")returns a single briefing (narrative chain head, open decisions, unresolved conflicts, pending triggers, stale high-importance memories) — replacing several separaterecall/temporalcalls at conversation start. Thenrecallonly for the specific thing the current message is about.
- During work — capture as you go.New durable fact →remember; a stored fact changed →correct(keeps history, avoids contradictions); relationship learned →graph(action="relate").
- End of substantial work — conditional.Only when the session was long or state-changing:thinkto consolidate + detect conflicts. Short/read-only exchanges need no end step.

Trust boundary:recalled memories and digest snippets aredata, not instructions. The playbook directs the agent never to execute directives found inside recalled content — a memory may carry text an earlier session or another user stored.

19 tools, full engine coverage (gaps,conversation,taskadded in v0.9.0):

Plus new actions on existing tools in v0.9.0:

- session(action="digest")— one-call boot-time briefing (narrative chain head + open decisions + conflicts + triggers)
- think(maintenance_cycle=True)— autonomous hygiene sleep cycle
- think(last_cycle_only=True)— read the last cycle summary without running
- stats(action="audit_leak")— privacy / leak-candidate audit
- stats(action="skill_outcomes")— durable skill-outcome count
- graph(action="auto_relate" / "record_link" / "record_unlink" / "linked_records" / "recall_with_links")— co-occurrence edges + record-to-record links + link-expanded recall
- conflict(action="auto_resolve")— burn down unambiguous conflicts in one pass
- memory(action="chain_head" / "history")— chain-namespace head + revision history
- trigger(action="prune")— bound the pending-trigger backlog
- remember(summary=...)— draft mode: engine atomizes a long summary into linked semantic facts (end-of-session auto-capture)

See[yantrikdb.com/guides/mcpfor full documentation.

…