context_analyze
Analyze context to extract entities and relationships
LABSCO SUMMARY
The context that dies with a session gets somewhere to live.
Work is saved under keys with categories and channels and read back filtered or diffed against a checkpoint, so "what did we decide about auth" has an answer that is not a scroll. context_prepare_compaction puts the critical items away before a compaction takes them, and a session can be branched to try a second approach and merged back when it works out.
What it is
A session-scoped context store: items saved under a key with a category, priority and channel, checkpoints over the whole session, and 38 tools for getting things back — filtered, searched, diffed against a point in time, or summarised.
What you get
- Sessions started against a project directory for git tracking, continued from an earlier one, branched to explore an alternative, and merged back under a conflict-resolution rule
- Items saved with category, priority, channel and a private flag, and batch save, update and delete as single atomic operations
- Named checkpoints that can carry cached files and git status, restored later by name or ID
- Critical context saved ahead of a compaction, and old context compressed toward a target size while named categories are preserved
- Retrieval that narrows: by key, key pattern, category, channel, priority and created-before/after, with sort, limit and offset
- Search within a session or across all of them, and natural-language search with a similarity threshold and topK
- A diff of what changed since a timestamp, a checkpoint or a relative time, and a timeline of activity grouped by period
- Entities and relationships extracted from saved context, related items walked to a depth you set, and visualization data for the graph
- A git commit created with the context saved alongside it, plus timestamped journal entries with tags
- Channels listed with counts and activity, and items reassigned between them with a dryRun first
- Session data exported and imported through the server's own exports directory
Requirements
npx on your PATH. No account and no key — context stays with the server, and exports are written to its exports directory.
Setup effort
One command — npx mcp-memory-keeper
USE CASES
01
Picking up yesterday's task without re-reading the thread
Start the session from the previous one so its items come with it, pull back the high-priority context for the category you are working in, and take a summary when the item list is longer than you want to read.
02
Carrying a decision through a context compaction
Save the critical items before the compaction takes them, and take a named checkpoint — with cached files and git status attached — that you can restore into a later session.
03
Working out what changed since the last checkpoint
Ask for the difference since a timestamp, a checkpoint or a relative time, and read the activity grouped by period when you want the shape of a week rather than a list of items.
04
Trying a second approach without contaminating the first
Branch the session, work the alternative in the branch, and merge it back with a conflict-resolution rule if it turns out better — or leave it unmerged if it does not.
05
Asking in your own words for something saved weeks ago
Natural-language search with a similarity floor finds the item when you cannot remember the key it was saved under, and the cross-session search widens it when the memory is from a different piece of work.
06
Tying a commit to the reasoning behind it
Commit through the server so the context is saved alongside the change, and add a journal entry when the reason is worth a sentence that no diff will carry.
38 tools
Counted once with only startup setup. Other versions, modes or optional credentials may expose a different list.
context_batch_delete
Delete multiple context items by keys or pattern in a single atomic operation
context_batch_save
Save multiple context items in a single atomic operation
context_batch_update
Update multiple context items with partial updates in a single atomic operation
context_branch_session
Create a branch from current session for exploring alternatives
context_cache_file
Cache file content with hash for change detection
context_channel_stats
Get detailed statistics for a specific channel or all channels
context_checkpoint
Create a named checkpoint of current context
context_compress
Intelligently compress old context to save space
context_delegate
Delegate complex analysis tasks to specialized agents
context_diff
Get changes to context items since a specific point in time (timestamp, checkpoint, or relative time)
context_export
Export session data (context items, cached files, and checkpoints with their links) for backup or sharing. JSON exports are written to the server's exports directory (<DATA_DIR>/exports, overridable via MEMORY_KEEPER_EXPORT_DIR); the response includes the full path, which can be passed back to context_import.
context_file_changed
Check if a file has changed since it was cached
context_find_related
Find entities related to a key or entity
context_get
Retrieve saved context by key, category, or session with enhanced filtering. Returns all accessible items (public items + own private items)
context_get_related
Get items related to a given context item
context_git_commit
Create git commit with automatic context save
context_import
Import previously exported session data. For security, imports are confined to the server's exports directory (where context_export writes); arbitrary filesystem paths are rejected.
context_integrate_tool
Track events from other MCP tools
context_journal_entry
Add a timestamped journal entry with optional tags and mood
context_link
Create a relationship between two context items
context_list_channels
List all channels with metadata (counts, activity, categories)
context_merge_sessions
Merge another session into the current one
context_prepare_compaction
Automatically save critical context before compaction
context_reassign_channel
Move context items between channels based on keys, patterns, or entire channel
context_restore_checkpoint
Restore context from a checkpoint
context_save
Save a context item with optional category, priority, and privacy setting
context_search
Search through saved context items with advanced filtering
context_search_all
Search across multiple or all sessions with pagination support
context_semantic_search
Search context using natural language queries
context_session_list
List recent sessions
context_session_start
Start a new context session with optional project directory for git tracking
context_set_project_dir
Set the project directory for git tracking in the current session
context_status
Get current context status and statistics
context_summarize
Get AI-friendly summary of session context
context_timeline
Get timeline of activities with optional grouping
context_visualize
Generate visualization data for the knowledge graph
context_watch
Create and manage watchers for real-time context change monitoring
BEFORE YOU INSTALL
What to know first
Tools you get
38 tools when we connected. TOOL_PROFILE=minimal exposes 8 tools and standard exposes 22; a config file at ~/.mcp-memory-keeper/config.json can define custom subsets. An unrecognised profile name falls back to the full set.
Requires
npx on PATH
CONTEXT COST
Moderate Standing cost
Rated by the size of the 38 tool definitions that ride along on every turn, not by how many there are
INSTALLATION
Connect your client
Server setup
TERMINAL
npx mcp-memory-keeperMCP Clients
Claude Code, Claude Desktop
JSON
{
"mcpServers": {
"memory-keeper": {
"command": "npx",
"args": [
"mcp-memory-keeper"
],
"env": {
"TOOL_PROFILE": "minimal"
}
}
}
}Claude Code — global install
CLAUDE CODE — GLOBAL INSTALL · TERMINAL
npm install -g mcp-memory-keeper
claude mcp add memory-keeper mcp-memory-keeper
