SmartMemory MCP server

Read from the package source without running it. The installed server may list more.

• agent_evaluation_get

  Get the current evaluation score for an agent on a given (dimension, domain) slot.

• agent_get_recall_profile

  Get an agent's recall profile.

• agent_set_recall_profile

  Set an agent's recall profile for personality-aware search re-ranking.

• clustering_runwrite action

  Run entity clustering/deduplication — merge duplicate entities by vector similarity.

• code_blame

  Which captured session authored this code? Given a git commit (or a file + line) in a local repo, find the Claude Code / Codex session whose edits structurally produced that code (CORE-CODE-PROVENANCE-1 Phase 2b).

• code_dead_code

  Find potentially dead (unreferenced) functions in an indexed codebase.

• code_dependencies

  Trace code dependencies -- what calls/imports/inherits what.

• code_index

  Index a Python codebase into SmartMemory's knowledge graph.

• code_read_transcript

  Read the conversation that authored a span of code (CORE-CODE-PROVENANCE-1 Phase 2c) — the *read* half of the blame->read chain. Copy the `code_read_transcript(...)` call that `code_blame` prints for a match: it renders a centered window of the authoring Claude Code / Codex transcript.

• code_search

  Search indexed code entities by name or description.

• decision_contradict

  Record contradicting evidence against a decision.

• decision_createwrite action

  Create a new decision with provenance tracking.

• decision_create_pendingwrite action

  Create a pending decision (GTM Acceptance Case) with unresolved requirements.

• decision_find_conflicts

  Find existing decisions that may conflict with this one.

• decision_get

  Retrieve a decision by ID.

• decision_list

  List active decisions with optional filtering.

• decision_provenance

  Get full provenance chain for a decision.

• decision_reinforce

  Record supporting evidence for a decision.

• decision_resolve_requirement

  Mark a requirement on a pending decision resolved by a memory item.

• decision_retract

  Retract a decision, marking it as no longer valid.

• decision_search

  Search for active decisions related to a topic.

• decision_supersede

  Replace a decision with a new one, marking the old as superseded.

• decision_try_activate

  Activate a pending decision if all requirements are resolved (no-op otherwise).

• decorator
• dev_load_context

  Load recent session context and related decisions for a work topic.

• dev_log_friction

  Log a friction event encountered during development.

• dev_query_decisions

  Search past development decisions by keyword, feature, or time range.

• dev_record_decision

  Record an architectural or design decision as a decision memory.

• dev_record_pattern

  Record a code pattern or convention as procedural memory.

• dev_save_session

  Save a development session summary as an episodic memory.

• evolution_dream

  **Deprecated:** CORE-MEMORY-DYNAMICS-1 M1b moved routing to at-ingest via the ConsolidationRouter. This tool is a no-op kept for caller compatibility.

• evolution_reinforce_opinionswrite action

  Run opinion reinforcement — update confidence scores based on new evidence.

• evolution_status

  Get status of memory evolution processes.

• evolution_synthesize_observationswrite action

  Run observation synthesis — create entity summaries from scattered facts.

• evolution_synthesize_opinionswrite action

  Run opinion synthesis — detect patterns in episodic memories and form opinions.

• evolution_trigger

  Trigger a memory evolution cycle — runs clustering, deduplication, and consolidation.

• get_working_context

  Retrieve a structured surfacing response (CORE-MEMORY-DYNAMICS-1 M1a).

• insight_feedbackwrite action

  Update memory system based on user feedback.

• insight_ground

  Ground a memory item to an external source URL for provenance.

• insight_health

  Get memory health summary — item counts, health score, and detected issues.

• insight_maintenance_status

  Get maintenance status — orphaned, stale, and low-confidence memory counts.

• insight_personalize

  Apply personalization settings — traits and preferences for memory retrieval.

• insight_plugins

  List available extractors and enrichers.

• insight_reflect

  Reflect on memory patterns — identify key concepts, top entities, and dominant topics.

• login

  Authenticate with a SmartMemory API key. Restart MCP server to unlock PRO tools.

• memory_add

  Add a memory item directly without the NLP pipeline.

• memory_anchor_check_drift

  Check if recent outputs have drifted from anchored requirements.

• memory_anchor_clear

  Clear (deactivate) anchors for a session.

• memory_anchor_graduate

  Graduate active anchors to persistent decisions.

• memory_anchor_list

  List active anchors for a session.

• memory_anchor_set

  Set a spec anchor to pin a requirement or constraint for the session.

• memory_auto

  Enable or disable automatic memory lifecycle.

• memory_check_failure

  Check for matching past failures before retrying.

• memory_clear

  Clear all memories permanently (requires confirm=True).

• memory_deletewrite action

  Delete a memory item by ID.

• memory_distill

  Ingest a conversation turn pair into pending memory for later recall.

• memory_explain

  Explain a memory's full provenance: origin, every belief the system held for it over time, what replaced it, what it derives from, and whether its history is cryptographically intact.

• memory_export

  Export all memories as an OKF bundle directory or tar.gz archive.

• memory_feedback

  Report which search results you used from a previous memory_search call.

• memory_get

  Retrieve a memory item by ID with full content and metadata.

• memory_get_violation_patterns

  Return the ``## Detection patterns`` catalog from local rule files.

• memory_import

  Losslessly import an OKF bundle through the direct add path.

• memory_ingest

  Ingest content through the full NLP pipeline with entity extraction.

• memory_ingest_conversation

  Ingest a conversation as session chunks with full pipeline processing (RLM-1g).

• memory_ingest_document

  Ingest a document from a URL or file path.

• memory_ingest_structured

  Ingest structured data via a registered handler, bypassing NLP pipeline.

• memory_list

  List memories with pagination, optionally filtered by exact metadata match.

• memory_log_failure

  Log an agent failure for cross-session error recovery.

• memory_migrate

  Migrate all memories to a different backend via a temporary OKF bundle.

• memory_plan_active

  Get all active plans.

• memory_plan_createwrite action

  Create a plan with optional task nodes.

• memory_plan_get

  Get a plan with all its tasks.

• memory_plan_update_taskwrite action

  Update a task's status within a plan (pending, in_progress, complete, blocked).

• memory_policy_bundle

  Compile active workspace decisions into a Stratum policy bundle.

• memory_recall

  Recall recent and relevant memories for the current context.

• memory_recall_pack

  Assemble the most useful context that fits in a token budget.

• memory_resolve_aliases

  Consolidate fragmented entity aliases over the workspace graph.

• memory_search

  Search memories using semantic similarity with optional hybrid mode.

• memory_search_advanced

  Advanced search using Similarity Graph Traversal (SSG) algorithms.

• memory_search_by_metadata

  Search memories by exact metadata key-value match.

• memory_stats

  Get memory count statistics grouped by type.

• memory_updatewrite action

  Update a memory item's properties (CORE-CRUD-UPDATE-1).

• pattern_get

  Fetch one pattern by id, or null if not found.

• pattern_list

  Return every active pattern (no filters).

• pattern_query

  Query the adherence pattern catalog.

• peer_chat

  Ask a question about a peer and get an answer synthesized from their memory.

• read_around

  Auto-centered conversation read (CORE-RECALL-CENTERED-1 Phase 2).

• reasoning_challenge

  Challenge an assertion against existing knowledge to detect contradictions.

• reasoning_extract_trace

  Extract reasoning traces from content — detects chain-of-thought patterns.

• reasoning_fuzzy_confidence

  Get multi-dimensional confidence score for a decision (evidence, recency, consensus, directness).

• reasoning_proof_tree

  Build an auditable proof tree for a decision, tracing evidence back to sources.

• reasoning_query

  Route a query to the cheapest effective retrieval method (symbolic, semantic, or hybrid).

• reasoning_query_traces

  Query stored reasoning traces — ask 'why' questions about past decisions.

• reasoning_resolve_conflict

  Resolve a conflict between assertions via Wikipedia, LLM, grounding, or recency.

• switch_team

  Switch this session to another workspace you belong to.

• transcript_search

  Search your own past Claude Code and Codex sessions by meaning (DIST-CC-INGEST-1 Phase 4). Use it to recover prior context — "have I hit this error before?", "why did we choose X?", "what did I try last time?" — instead of re-deriving it.

• transcript_status

  Report how much of your Claude Code / Codex history is actually searchable (DIST-CC-INGEST-1 Phase 4), and whether an import is running right now.

• whoami

  Show the current hosted session: user, workspace, and auth kind.

• zettel_backlinks

  Get notes that link TO this note (backlinks).

• zettel_connections

  Get all connections (backlinks + forward links) for a note.

• Code scan47 source files scanned25/25
• –Live reliabilityno gateway calls yet and no remote to proben/a
• –Tool poisoningtools not inspected (local package is not executed); not countedn/a
• Auth qualitylocal package, no credentials required12/15
• Maintenanceno repository listed3/15
• Maintainer identityno repository or website to verify2/10

SmartMemory: common questions

Is SmartMemory MCP server safe?

With care: it is graded C, so read the findings first (65/100). Read the SmartMemory safety report

How do I install SmartMemory?

It runs on your machine. Copy the Claude Code, Claude Desktop or Cursor config from the install section.

Does SmartMemory need an API key?

Not as far as the registry entry and our scan can tell: no credentials are declared or required.

Is SmartMemory maintained?

The latest release is v1.4.94.

What can I use instead of SmartMemory?

Servers from other publishers that do the same job: Total Agent Memory MCP server, Fidelis Memory MCP server and rag-rat MCP server. Compare all SmartMemory alternatives.