Skip to main content
3ngram’s MCP server provides 18 tools for AI assistants to interact with the knowledge base.

​
Tool catalog

ToolPurposeR/W
rememberPersist a structured memory with type, scope, project, due dateWrite
recallRetrieve memories by semantic similarityRead
get_memoriesFetch full memory content by IDsRead
resolveMark a memory as resolved with optional noteWrite
commitmentsList open commitmentsRead
blockersList open blockersRead
overdueList commitments past their deadlineRead
stale_commitmentsList commitments with no activity for N+ daysRead
searchHybrid semantic + BM25 search across content chunksRead
reindexRe-index markdown content into content chunksWrite
reclassify_memoryChange the type of a memoryWrite
suggest_reclassificationsFind context memories that may be misclassifiedRead
triage_staleSurface stale memories with resolution promptsRead
archive_memoriesArchive resolved or stale memories by IDWrite
handoffExport session context as structured JSONRead
configure_scopeAssign a project to a scope (work/personal)Write
reapply_scope_mappingsReapply scope mappings to existing memoriesWrite
statusIndex statistics and memory breakdownRead

​
Key parameters

​
remember

ParameterTypeRequiredDescription
topicstringYesShort identifier for the memory
contentstringYesFull memory content
memory_typestringNodecision, commitment, blocker, preference, pattern, context
projectstringNoProject tag
scopestringNowork or personal
due_datestringNoISO date for commitments
statusstringNoopen, waiting, scheduled

​
recall

ParameterTypeRequiredDescription
topicstringYesSearch query
briefbooleanNoCompact one-line-per-memory format
limitintegerNoMax results (default 10)
memory_typestringNoFilter by type
projectstringNoFilter by project
scopestringNoFilter by scope
sourcestringNoFilter by source (manual, hook, auto)

​
resolve

ParameterTypeRequiredDescription
memory_idintegerYesMemory to resolve
resolution_notestringNoWhat happened

​
Output format

All tools return human-readable formatted strings, not raw data. Example commitments output: 
Open Commitments (3):
  [122] Implement memory consolidation UI (5 days) [project: engram]
  [125] Write integration tests for OAuth flow (3 days) [project: engram]
  [131] Review Cursor MCP configuration (1 day) [project: climbr]

​
Progressive disclosure

Use recall(brief=True) to scan, then get_memories(ids=[...]) to read details. This keeps token usage low when browsing large memory sets.

​
Auto-consolidation

When remember finds an existing memory with 85%+ similarity, it merges instead of creating a duplicate. Consolidation uses advisory locks to prevent race conditions.