Context Mem
Persistent memory layer that solves the forget-versus-context-explosion problem in AI coding sessions. Every tool call gets ingested, summarized through 14 content-aware compressors, and written into a local SQLite store with a synced markdown vault you can browse in Obsidian. Raw outputs compress 99% (365 KB to 3.2 KB tested) while maintaining 98%+ retrieval recall through hybrid BM25 plus vector search with optional LLM reranking. Exposes 44 tools for querying entities, topics, sessions, and knowledge pages. Auto-configures for Claude Code, Cursor, Windsurf, Cline, Continue, and others with one init command. Fully local, no API keys. Reference implementation of Karpathy's LLM Wiki pattern with the observation pipeline built in.
- GitHub stars
- 13
- tools
- 40
- transport
- STDIO
- last updated
- Apr 15, 2026
Tools
40 toolsPublic tool metadata for what this MCP can expose to an agent.
observeStore observation with auto-summarization, importance scoring, entity extraction, topic detection
Store observation with auto-summarization, importance scoring, entity extraction, topic detection
No parameter schema in public metadata yet.
recallRetrieve verbatim content by filter (importance, type, flag, time)
Retrieve verbatim content by filter (importance, type, flag, time)
No parameter schema in public metadata yet.
searchHybrid search (BM25 + vector + optional LLM judge)
Hybrid search (BM25 + vector + optional LLM judge)
No parameter schema in public metadata yet.
askNatural language Q&A over the full memory store
Natural language Q&A over the full memory store
No parameter schema in public metadata yet.
timelineReverse-chronological observations with importance badges and flags
Reverse-chronological observations with importance badges and flags
No parameter schema in public metadata yet.
save_knowledgeSave a knowledge entry with contradiction detection + temporal validity windows
Save a knowledge entry with contradiction detection + temporal validity windows
No parameter schema in public metadata yet.
search_knowledgeSearch (superseded entries filtered by default)
Search (superseded entries filtered by default)
No parameter schema in public metadata yet.
promote_knowledgePromote to global cross-project store
Promote to global cross-project store
No parameter schema in public metadata yet.
global_searchSearch across all projects simultaneously
Search across all projects simultaneously
No parameter schema in public metadata yet.
resolve_contradictionResolve knowledge conflicts (supersede / merge / keep / archive)
Resolve knowledge conflicts (supersede / merge / keep / archive)
No parameter schema in public metadata yet.
merge_suggestionsView cross-project duplicate suggestions
View cross-project duplicate suggestions
No parameter schema in public metadata yet.
graph_queryTraverse entity relationships
Traverse entity relationships
No parameter schema in public metadata yet.
add_relationshipLink entities with typed relationships
Link entities with typed relationships
No parameter schema in public metadata yet.
graph_neighborsFind connected entities (configurable depth)
Find connected entities (configurable depth)
No parameter schema in public metadata yet.
temporal_queryQuery what was true at a specific point in time
Query what was true at a specific point in time
No parameter schema in public metadata yet.
time_travelCompare project state at two arbitrary timestamps
Compare project state at two arbitrary timestamps
No parameter schema in public metadata yet.
explain_decisionWalk evidence chain backward to reconstruct why a decision was made
Walk evidence chain backward to reconstruct why a decision was made
No parameter schema in public metadata yet.
predict_lossIdentify observations at risk of compression/deletion
Identify observations at risk of compression/deletion
No parameter schema in public metadata yet.
generate_storyGenerate PR description, standup update, ADR, or onboarding guide
Generate PR description, standup update, ADR, or onboarding guide
No parameter schema in public metadata yet.
entity_detectDetect entities in arbitrary text
Detect entities in arbitrary text
No parameter schema in public metadata yet.
find_tunnelsFind cross-project topic connections
Find cross-project topic connections
No parameter schema in public metadata yet.
wake_upToken-budgeted context primer for session start
Token-budgeted context primer for session start
No parameter schema in public metadata yet.
restore_sessionRestore session from checkpoint
Restore session from checkpoint
No parameter schema in public metadata yet.
handoff_sessionCross-session continuity package
Cross-session continuity package
No parameter schema in public metadata yet.
agent_registerRegister an agent with role and capabilities
Register an agent with role and capabilities
No parameter schema in public metadata yet.
agent_statusCheck all active agents and their claimed resources
Check all active agents and their claimed resources
No parameter schema in public metadata yet.
claim_filesClaim files to prevent parallel-agent conflicts
Claim files to prevent parallel-agent conflicts
No parameter schema in public metadata yet.
agent_broadcastBroadcast a finding to all agents in the project
Broadcast a finding to all agents in the project
No parameter schema in public metadata yet.
configureUpdate runtime configuration
Update runtime configuration
No parameter schema in public metadata yet.
budget_statusToken budget management
Token budget management
No parameter schema in public metadata yet.
budget_configureToken budget management
Token budget management
No parameter schema in public metadata yet.
summarizeSummarize content without storing (one-shot)
Summarize content without storing (one-shot)
No parameter schema in public metadata yet.
executeRun code (JS, TS, Python, Shell, Ruby, Go, Rust, PHP, Perl, R, Elixir)
Run code (JS, TS, Python, Shell, Ruby, Go, Rust, PHP, Perl, R, Elixir)
No parameter schema in public metadata yet.
index_contentIndex with code-aware chunking
Index with code-aware chunking
No parameter schema in public metadata yet.
search_contentSearch indexed chunks
Search indexed chunks
No parameter schema in public metadata yet.
list_peopleBrowse entities and topics
Browse entities and topics
No parameter schema in public metadata yet.
list_topicsBrowse entities and topics
Browse entities and topics
No parameter schema in public metadata yet.
import_conversationsImport conversation history
Import conversation history
No parameter schema in public metadata yet.
browseRetrieve observations by person, entity, or topic
Retrieve observations by person, entity, or topic
No parameter schema in public metadata yet.
diagnosticsError log, pipeline stats, storage health
Error log, pipeline stats, storage health
No parameter schema in public metadata yet.
Sponsors
Agent CaffeineSponsoredClose your lid, come back to finished work
SnugSponsoredThe clipboard manager for the AI eraPromote your product hereReach 125k+ visitors monthly
WaxSponsoredBlock anything with supervision, without erasing your iPhonePromote your product hereReach 125k+ visitors monthly
knobsSponsoredSee your site on any phone, any day, in any languageReadme
Context Mem
Memory + context infrastructure for AI agents. Remembers everything. Compresses everything. Fully local.
-gold){kind=icon}
The Problem
Two problems with today's AI tooling that no one has solved together in a single package.
Details
- Repository
- github.com/jubakitiashvili/context-mem
- Categories
- AI & LLM Tools, Documents & Knowledge
- Package
context-mem- Transport
- STDIO
- Registry
- MCP registry, active
Related AI & LLM Tools MCP servers
View all
Synapse Layer — Trust Infrastructure for AI Agentssynapselayer/synapse-layerMCP-native Trust Infrastructure for AI Agents. Persistent encrypted memory with Trust Quotient.13stars
Hashnethashgraph-online/hashnet-mcp-jsUniversal AI Agent Discovery - search 59,000+ agents across MCP, NANDA, A2A, OpenRouter0stars
Alaya Mcpsecurityronin/alayaLocal memory engine for AI agents with knowledge graphs, forgetting, and semantic recall12stars
ChromaDB Remote MCP Servermeloncafe/chromadb-remote-mcpRemote ChromaDB vector database MCP server with streamable HTTP transport12stars
Pindocvar-gg/pindocCode-pinned team memory for AI coding agents — typed artifacts, MCP-native, self-host.12stars
Velociraghaseebkhalid1507/velociragLightning-fast RAG for AI agents. 4-layer fusion, ONNX Runtime, sub-200ms search.10stars