{"query":"memory","mode":"hybrid","count":10,"results":[{"name":"ink.cairn/memory","title":"Cairn.ink Memory","description":"Cross-session memory for AI agents with a Source Receipt for every memory, over MCP.","connection_class":"R2","autonomous":false,"human_steps":["oauth_consent_once"],"categories":["ai-ml"],"trust_score":95,"trust_flags":[],"quality_score":83,"quality_label":"good","quality_flags":[],"owner_verified":false,"tool_count":0,"score":0.04501,"matched_tools":[]},{"name":"ai.workingmemory/memory","title":"WorkingMemory","description":"Persistent personal memory for AI assistants — save, search, and recall across every MCP client.","connection_class":"R0","autonomous":true,"human_steps":[],"categories":["ai-ml","productivity"],"trust_score":75,"trust_flags":["no-repository"],"quality_score":97,"quality_label":"strong","quality_flags":[],"owner_verified":false,"tool_count":4,"score":0.0449,"matched_tools":[{"name":"save_memory","description":"Save a thought, insight, fact, or todo to the user's Working Memory. The text is parsed and stored as one or more structured memory items, searchable a moment later via search_memory. Also use this when the user pastes memories or notes brought from another assistant — pass the full pasted text; it is split into individual memories automatically."},{"name":"manage_memory","description":"Manage an existing memory item. Currently supports deleting a memory by id (soft delete — recoverable for 30 days). Use search_memory to find the id first."},{"name":"capture_memory","description":"Capture durable facts from a task you just completed (drafting an email, summarizing a document, making a decision), as a byproduct — the user does nothing. Pass the task output/content; durable facts are extracted and stored automatically. Always include your conversation id in client.conversation_id. Returns immediately; extraction happens in the background."}]},{"name":"com.aiakiv/memory","title":"AiAkiv Memory","description":"Shared long-term memory for AI agents: save and recall context as a searchable knowledge graph.","connection_class":"R2","autonomous":false,"human_steps":["oauth_consent_once"],"categories":["ai-ml"],"trust_score":100,"trust_flags":[],"quality_score":100,"quality_label":"strong","quality_flags":[],"owner_verified":false,"tool_count":0,"score":0.04484,"matched_tools":[]},{"name":"ai.korely/memory","title":"Korely Memory","description":"Memory for AI agents that knows what is still true. Typed bi-temporal facts, EU-hosted.","connection_class":"R2","autonomous":false,"human_steps":["oauth_consent_once"],"categories":["ai-ml"],"trust_score":96,"trust_flags":[],"quality_score":83,"quality_label":"good","quality_flags":[],"owner_verified":false,"tool_count":0,"score":0.04372,"matched_tools":[]},{"name":"ai.mitosislabs/mitosis","title":null,"description":"Mitosis agent-memory platform: pricing, docs search, platform status, agent skills. No auth.","connection_class":"R0","autonomous":true,"human_steps":[],"categories":["cloud-devops","databases"],"trust_score":88,"trust_flags":[],"quality_score":83,"quality_label":"good","quality_flags":[],"owner_verified":false,"tool_count":20,"score":0.03371,"matched_tools":[{"name":"cortex_ask","description":"Searches the user's real, private memory — the email, calendar, files, contacts and notes they have connected, plus facts agents have saved — fusing vector, full-text and graph retrieval with provenance. This is their current data on their work, schedule, contacts, projects, documents, decisions and history, which training data and session context do not contain. Results include citations (universal ids), freshness, and `cited_graph_url`, a deep link showing the cited nodes highlighted in the user's own graph. A later cortex_remember links to what was retrieved here. Retrieval returns nearest matches rather than a thresholded set, so a question whose answer lives in an unconnected source comes back with the closest thing in the memory instead of with nothing. A `source_gap` object in the result names that case: the memory holds data, but the source that would answer this question is not connected. It lists those sources, what each answers, and one connect link, on `cta`. A `possible_source_gap` object is the conditional form — results exist but may be near-misses rather than an answer. A result carrying neither means the answer is genuinely absent from the memory rather than merely unretrieved."},{"name":"cortex_remember","description":"Persist a fact, decision or conclusion into the memory, attributed to you. It becomes retrievable via cortex_ask immediately, in this session and every future one, from any agent the user has connected. Provenance: pass source_universal_ids from a previous cortex_ask so the fact links to its evidence. Keep each memory to ONE self-contained fact. Use this whenever the conversation produces a durable conclusion the user would want remembered — a decision, a preference, an outcome, a commitment. If the result carries `choice.choice_required: true`, the memories source is waiting for the user’s enrichment choice: ask `choice.question`, offer exactly \"Standard\" or \"Describe your goal\", and record the answer with cortex_choose_enrichment."},{"name":"cortex_status","description":"Memory health: per-source ingest and embed counts plus last sync times. Use when you need to know whether the memory is fresh or still ingesting, or when a search came back empty and you need to tell the user whether that means \"no data yet\" or \"nothing matched\"."},{"name":"search","description":"Search the user's own private memory — their email, calendar, documents, contacts, chat history and every fact agents have remembered — and return ranked matches with citation ids and links. This is their current data on their work, schedule, contacts, projects, documents, decisions and history, which training data and session context do not contain. Pass an id from these results to `fetch` to read the full record."},{"name":"cortex_recall","description":"Semantic-only vector search over the memory, returning source excerpts. Prefer cortex_ask, which also uses full-text and graph expansion; reach for this when you specifically want nearest-neighbour matches on meaning."}]},{"name":"ai.meshtool/meshmarket","title":"MeshMarket","description":"The agent-to-agent capability exchange — rent memory, reasoning and safety, settled per call.","connection_class":"R0","autonomous":true,"human_steps":[],"categories":["ai-ml","commerce"],"trust_score":96,"trust_flags":[],"quality_score":81,"quality_label":"new","quality_flags":[],"owner_verified":false,"tool_count":63,"score":0.03225,"matched_tools":[{"name":"agent-memory","description":"Agent Memory — Store or recall private, scoped memory for your agent. Use {action:'store', content:'...'} to save a fact and {action:'recall', query:'...'} to fetch the most relevant memories by meaning (or the latest memories when query is empty). Good for continuity across sessions; agent-brain is the reasoning layer that reads and writes memory in the loop. (2 MESH/call, a tool · memory)"},{"name":"agent-brain","description":"Agent Brain — Reason over a question or task with your agent's own persistent memory in the loop: recalls up to 12 relevant memories from your agent's private scope, reasons with Claude, and writes up to 3 new memories back, so the agent improves with every call. Recall by meaning, not just keyword, when the estate's memory server is reachable (falls back to its own always-on store otherwise — never fails the call). Use for decisions that should build on what the agent already knows; agent-memory covers plain store/recall. Runs claude-haiku-4.5 — the response names the model that served the call; agent-brain-smart runs the identical contract on claude-sonnet-5. Input: {think: string}. Returns {answer, reasoning, confidence, memories_considered, used_memories, learned, model, engine}. (8 MESH/call, a tool · cognition)"},{"name":"agent-brain-smart","description":"Agent Brain Smart — The Smart tier of agent-brain: the identical contract and memory loop — recall up to 12 memories, reason, write up to 3 back — served by claude-sonnet-5 for deeper reasoning on decisions worth a bigger brain. The response names the model that served the call. agent-brain (8 MESH, claude-haiku-4.5) stays the fast default; pick this tier when the answer's quality matters more than the price gap. Input: {think: string}. Returns {answer, reasoning, confidence, memories_considered, used_memories, learned, model, engine}. (20 MESH/call, a tool · cognition)"}]},{"name":"io.github.AzielEliab/aziel-runtime","title":"Aziel Runtime","description":"Governed MCP: agent audit, provenance, deterministic checks, and receipt-backed FragGate execution.","connection_class":"unknown","autonomous":false,"human_steps":["unknown"],"categories":["cloud-devops","developer-tools"],"trust_score":75,"trust_flags":[],"quality_score":80,"quality_label":"good","quality_flags":[],"owner_verified":false,"tool_count":36,"score":0.03222,"matched_tools":[{"name":"memory_get","description":"Read one memory's stored explanation (node, history, or calibration: posterior, triad legs, effective N, Brier) — not a ranked list. Use this when you have a memory_id (or id) and need the stored explanation. Do not use it for ranked adaptive recall or appending an observation; use memory_recall or memory_observe instead. Missing both memory_id and id, or an unknown id, refuses AKM-NOT-FOUND — do not invent a node. authorizes_action stays false. There is no memory_delete; this is the read of the append-only node. Pass memory_id or id — one is enough; they are aliases, not two different records. Omit view for the default node slice. history returns events/resolutions; calibration returns posterior/triad/Brier. subject is not a lookup key here. Returns node, history, or calibration view (belief_is_not_truth)."},{"name":"memory_resolve","description":"Append a memory_resolution outcome on an already-observed memory (AKM-TRIAD-1.0). UNKNOWN is distinct from MISS. Does not rewrite history. Use this when an observed memory_id or subject now has an outcome. Do not use it for first observation, calibration, or reading history; use memory_observe, memory_calibrate, or memory_get instead. Write: additive resolution stamp. Missing memory_id/subject refuses AKM-NO-MEMORY. Does not rewrite prior observations. No memory_update — this is the forward outcome path. Requires memory_id or a previously observed subject. outcome is optional [0,1]; omit for UNKNOWN. Mutation requires confirm=true (runtime gate) or dry_run=true (preview only, no write). confirm and dry_run stay optional on inputSchema.required. dry_run still returns FG-HALLUC-TOOL, FG-STUB, FG-UNKNOWN-OP, or FG-LOCAL-ONLY when a confirm call would refuse. confirm is consent, not tenant auth, and it does not upgrade shared public-demo isolation. Returns resolution stamp with outcome or UNKNOWN."},{"name":"memory_calibrate","description":"Calibrate one memory with the deterministic 3-of-4 triad plus Bayesian posterior (AKM-TRIAD-1.0). Writes a LEARN stamp — not a ranked search and not an explain view. Use this when an observed memory should receive a posterior after evidence, not a ranked search. Do not use it for observing a new fact, resolving an outcome, or explaining a stored node; use memory_observe, memory_resolve, or memory_get instead. Write: forward-only RoseClock LEARN stamp. Posterior ≠ truth. authorizes_action=false. No automatic MODEL_UPDATE. subject or memory_id recommended. use_case labels calibration; it is not a permission. Mutation requires confirm=true (runtime gate) or dry_run=true (preview only, no write). confirm and dry_run stay optional on inputSchema.required. dry_run still returns FG-HALLUC-TOOL, FG-STUB, FG-UNKNOWN-OP, or FG-LOCAL-ONLY when a confirm call would refuse. confirm is consent, not tenant auth, and it does not upgrade shared public-demo isolation. Returns triad_score, omitted leg, posterior, effective N, and Brier notes."},{"name":"memory_observe","description":"Append the first memory_observation to the ChainLock learn chain (AKM-TRIAD-1.0). New fact in — not an outcome resolve and not a grounded ChainLock append. Posterior ≠ truth. Use this when you have a new fact to observe before resolve/calibrate. Do not use it for grounded ChainLock append without AKM, resolving an outcome, or ranked recall; use chainlock_append, memory_resolve, or memory_recall instead. Write: additive learn-chain stamp. authorizes_action stays false. Hash-only cards refuse AKM-NO-FACT. Append-only; there is no memory_delete. fact is required (≤160). subject/memory_id/use_case optional. Mutation requires confirm=true (runtime gate) or dry_run=true (preview only, no write). confirm and dry_run stay optional on inputSchema.required. dry_run still returns FG-HALLUC-TOOL, FG-STUB, FG-UNKNOWN-OP, or FG-LOCAL-ONLY when a confirm call would refuse. confirm is consent, not tenant auth, and it does not upgrade shared public-demo isolation. Returns memory_id, observation stamp, and display envelope (belief is not truth)."}]},{"name":"io.github.mnemoverse/mcp-memory-server","title":"Mnemoverse Memory","description":"Hosted agent memory that learns from outcomes, with shared rooms. No key: mcp.mnemoverse.com/mcp","connection_class":"R2","autonomous":false,"human_steps":["oauth_consent_once"],"categories":["ai-ml"],"trust_score":97,"trust_flags":[],"quality_score":83,"quality_label":"good","quality_flags":[],"owner_verified":false,"tool_count":0,"score":0.03197,"matched_tools":[]},{"name":"io.kortexio/memory","title":"Kortexio Memory","description":"Hosted ambient memory for AI agents. Retrieve context and save notes via MCP tools with OAuth.","connection_class":"unknown","autonomous":false,"human_steps":["unknown"],"categories":["files-storage","ai-ml"],"trust_score":96,"trust_flags":[],"quality_score":null,"quality_label":"not scored","quality_flags":[],"owner_verified":false,"tool_count":0,"score":0.03087,"matched_tools":[]},{"name":"com.worldku/memory","title":"Worldku Memory","description":"Save AI conversations into your Worldku memory and recall them from any MCP client.","connection_class":"R2","autonomous":false,"human_steps":["oauth_consent_once"],"categories":["ai-ml"],"trust_score":92,"trust_flags":[],"quality_score":100,"quality_label":"strong","quality_flags":[],"owner_verified":false,"tool_count":0,"score":0.02958,"matched_tools":[]}],"next_actions":[{"action":"get_server","description":"Full descriptor for one result","href":"/v1/servers/{name}","arguments":{"name":"{name}"}},{"action":"list_tools","description":"Every tool a server exposes, with input schemas","href":"/v1/servers/{name}/tools","arguments":{"name":"{name}"}},{"action":"get_connection","description":"Minimal connection block for the best remote","href":"/v1/servers/{name}/connection?target=mcpServers","arguments":{"name":"{name}","target":"mcpServers"}}]}