{"domain":"memory.studiomeyer.io","count":1,"changes":[{"captured_at":"2026-05-18T14:28:24","card_hash":"315cce1dbfa56f57f8bc507a7e8121aa597c44134d890a3076e5a75120e76bc5","previous_card_hash":null,"diff":{"skills_added":[{"id":"analytics-insights","name":"Analytics, insights, contradictions, decay","description":"Knowledge graph health analytics, contradiction detection, learning decay (forgetting curve), proactive suggestions, communities clustering, deduplication, synthesis from learning clusters.","tags":["analytics","insights","health","communities"],"inputModes":null,"outputModes":null},{"id":"knowledge-graph","name":"Knowledge graph entity tracking","description":"Bi-temporal knowledge graph with entities, observations, relations. Time-travel queries via asOf parameter. Entity merge, deep-search, history, communities. 56 tools total in MCP surface.","tags":["knowledge-graph","entities","bi-temporal","observations"],"inputModes":null,"outputModes":null},{"id":"learn-and-decide","name":"Persist learnings and decisions","description":"Store typed learnings (pattern, mistake, insight, research, architecture, infrastructure, tool, workflow, performance, security) with built-in duplicate-prevention gatekeeper. Log decisions with confidence scores and alternatives.","tags":["memory","learn","decide","gatekeeper"],"inputModes":null,"outputModes":null},{"id":"memory-import","name":"Import ChatGPT, Claude, Gemini history","description":"Bulk import conversation history from ChatGPT, Claude.ai, Gemini exports. Auto-extracts learnings, entities, and relations into the knowledge graph.","tags":["import","chatgpt","claude","gemini","migration"],"inputModes":null,"outputModes":null},{"id":"memory-search","name":"Search across all memory","description":"Unified search with temporal decay across decisions, learnings, sessions, skills, evolution, entity observations. Trigram + full-text + LIKE. Recent memories rank higher.","tags":["memory","search","recall","rrf","temporal"],"inputModes":null,"outputModes":null},{"id":"multi-llm-memory","name":"Provider-agnostic memory backend","description":"Works with any LLM that speaks MCP: Claude (Anthropic), GPT (OpenAI), Gemini (Google), local models (Ollama, LM Studio), n8n, ChatGPT custom GPTs. Bring your own LLM, we provide the storage and search layer.","tags":["multi-llm","agnostic","mcp","byo-llm"],"inputModes":null,"outputModes":null},{"id":"session-continuity","name":"Session start, end, replay","description":"Auto-loaded context on session start, structured summaries on session end, full session replay with timeline visualization. Sessions persist across multiple AI conversations.","tags":["sessions","continuity","replay","timeline"],"inputModes":null,"outputModes":null},{"id":"skill-tracking","name":"Track which patterns work","description":"Skill registry with success/failure tracking per domain. Recipe lookup returns the best tool-chain for a task based on past experience.","tags":["skills","patterns","recipes"],"inputModes":null,"outputModes":null},{"id":"tenant-isolation","name":"Multi-tenant with Postgres RLS","description":"Shared-tables architecture with tenant_id and Postgres Row-Level-Security. EU Frankfurt (Supabase), DSGVO-compliant, Article 17 erasure supported. Bearer API key + OAuth 2.1 + Magic Link auth.","tags":["multi-tenant","rls","gdpr","oauth-2.1","eu"],"inputModes":null,"outputModes":null},{"id":"visualization-graph","name":"Interactive 3D knowledge graph","description":"Cinema-mode 3D-Force-Graph rendering of the knowledge graph with time-travel slider and session-replay timeline. Available as MCP Apps UI resources for clients that support sandboxed iframes (Claude.ai, VS Code Copilot, Goose, Postman, MCPJam).","tags":["visualization","3d","mcp-apps","time-travel"],"inputModes":null,"outputModes":null}],"skills_removed":[],"skills_changed":[],"fields_changed":[{"field":"name","before":null,"after":"StudioMeyer Memory"},{"field":"description","before":null,"after":"Long-term AI memory with bi-temporal knowledge graph. 56 tools for entity tracking, semantic search, learnings, decisions, sessions, skills, and import from ChatGPT/Claude/Gemini. Provider-agnostic — works with any LLM that speaks MCP. EU Frankfurt, DSGVO-compliant, multi-tenant on Supabase with Postgres RLS."},{"field":"version","before":null,"after":"3.16.13"},{"field":"protocolVersion","before":null,"after":"0.3.0"},{"field":"url","before":null,"after":"https://memory.studiomeyer.io/mcp"},{"field":"documentationUrl","before":null,"after":"https://studiomeyer.io/services/memory"},{"field":"iconUrl","before":null,"after":"https://memory.studiomeyer.io/favicon.ico"},{"field":"preferredTransport","before":null,"after":"JSONRPC"}],"other_changed":true,"is_empty":false,"human_summary":"added 10 skills · name ∅ → StudioMeyer Memory · description ∅ → Long-term AI memory with bi-temporal kno · version ∅ → 3.16.13 · protocolVersion ∅ → 0.3.0 · url ∅ → https://memory.studiomeyer.io/mcp · documentationUrl ∅ → https://studiomeyer.io/services/memory · iconUrl ∅ → https://memory.studiomeyer.io/favicon.ic · preferredTransport ∅ → JSONRPC"}}]}