Graph-based memory for Claude Code, Cursor & Devin. SUMA replaces flat files with a mathematical knowledge graph — relationships, emotions, patterns — not just keywords.
{
"mcpServers": {
"suma-memory": {
"url": "https://sumapro-mcp.quadframe.work/mcp",
"headers": {
"Authorization": "Bearer sk_live_..."
}
}
}
}SUMA doesn't just store text. It builds a Knowledge-Weighted Intent Layer (K-WIL) using Connective Topology and Behavioral Signal Detection to isolate the exact context your AI needs.
Finds connections flat-file search can't. Ask "Who knows my deployment pipeline?" — SUMA traces the graph and finds the answer.
Detects intensity in text — life events, urgency, frustration — and prioritizes important context accordingly.
Use Claude Monday, Cursor Tuesday, Devin Wednesday — all share the same SUMA brain. One memory, every tool.
Patented formulas calculate what matters most — not just recency or frequency. Context that counts rises to the top.
Finds contradictions in your knowledge graph. "You said X last week but Y today." Resolves conflicts automatically.
Detects repeated behaviors over time. "You always deploy on Fridays." Cross-pattern causation analysis.
Subscribe at $4.99/month. We email you a sk_live_... key instantly.
Works with any MCP-compatible tool.
// Claude Code, Cursor, or any MCP client
{
"mcpServers": {
"suma-memory": {
"url": "https://sumapro-mcp.quadframe.work/mcp",
"headers": { "Authorization": "Bearer sk_live_..." }
}
}
}SUMA automatically ingests context, builds the knowledge graph, and serves targeted results — 200 tokens instead of 15,000.
Your AI calls these tools automatically via the Model Context Protocol.
suma_search
Weighted graph search with emotional context
// AI calls this automatically
suma_search(
query: "deployment pipeline",
depth: 2, // multi-hop traversal
limit: 5, // top 5 results
sphere: "work" // scope to work context
)
// Returns: nodes + confidence scores + emotional signals + graph pathssuma_ingest
Add knowledge to the graph
suma_ingest(
text: "The API uses PostgreSQL with org_id isolation",
sphere: "architecture",
extract_relationships: true // auto-extracts triplets
)
// Automatically creates: API → uses → PostgreSQL, PostgreSQL → has → org_id_isolationsuma_talk
Bidirectional — searches AND learns in one call
suma_talk(
message: "We're migrating to Cloud Run next sprint",
persona: "companion"
)
// Returns: relevant context + learned nodes + emotional analysissuma_node
Get a node with its full weight profile
suma_node(
node_id: "POSTGRESQL",
include_neighbors: true,
include_weight_profile: true
)
// Returns: content, sphere, neighbors, harmonic weight, emotional intensitysuma_patterns
Detect behavioral patterns and causation
suma_patterns(
query: "deployment",
include_cross_patterns: true
)
// Returns: temporal patterns, cross-pattern causation
// e.g., "deploy_friday" CAUSES "hotfix_monday"Free to try. $29/month for the full brain.
| Feature | Claude memory.md | Cursor .cursorrules | Mem0 | SUMA Pro |
|---|---|---|---|---|
| Graph relationships | ✗ | ✗ | ✗ | ✓ |
| Emotional detection | ✗ | ✗ | ✗ | ✓ |
| Multi-hop traversal | ✗ | ✗ | ✗ | ✓ |
| Cross-tool sharing | ✗ | ✗ | ~ | ✓ |
| Paradox detection | ✗ | ✗ | ✗ | ✓ |
| Pattern detection | ✗ | ✗ | ✗ | ✓ |
| Patented algorithms | ✗ | ✗ | ✗ | 5 US patents filed |