Atlas / MCP servers / varun29ankus / Shodh-Memory

Shodh-MemoryBLOCK

Name: Shodh-Memory
Author: varun29ankus

mcp/varun29ankus/shodh-memory

MCP server by varun29ankus · agent memory · agentic ai · ai agents · Audited 2026-10-06

Local, LLM-free memory for AI agents. A single offline Rust binary — deterministic and auditable — that learns from use, forgets the irrelevant, and strengthens what matters. No cloud, no API keys.

Verdict

BLOCK

Grade

Trust score

54 /100

Exposed tools

67 47r · 12w · 8d

Transport

stdio · streamable-http

License

Apache-2.0

Stars

301

Overview

From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.

Shodh-Memory

Persistent cognitive memory for AI agents and robots — with no LLM in the loop. Remembers what matters, forgets what doesn't, gets smarter with use.

Read from source at commit fda2ad7f58a3OBSERVED · 2026-10-06

Connect

Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control. Replace the environment placeholders with a token scoped to the least it needs.

claude-code (npm)

claude mcp add memory-mcp --env SHODH_API_KEY=${SHODH_API_KEY} -- npx -y @shodh/[email protected]

Exposed tools (67)

47 read · 12 write · 8 destructive. Blast radius: 8 tools can delete or overwrite — an agent that can be talked into calling a tool can be talked into calling this one.

Tool	Risk	Description
`add_causal_link`	write	Record an explicit causal edge between two memories. from_memory_id is the cause/origin (the earlier event), to_memory_id is the effect (the later one); the relation reads from→to, e.g. relation=Caused means
`add_project`	write	Create a new project to group todos. Use parent to create a sub-project under another project.
`add_todo`	write	Add a task to your todo list. Supports GTD workflow with projects, contexts (@computer, @phone), priorities, due dates, and subtasks (via parent_id).
`add_todo_comment`	write	Add a comment to a todo. Use to track progress, notes, or resolution details.
`archive_project`	read	Archive a project. Archived projects are hidden by default but can be restored.
`backup_create`	write	Create a backup of all memories. Returns backup metadata including ID, size, and checksum. Backups are stored locally and can be restored later.
`backup_list`	read	List all available backups for this user. Returns backup history with IDs, timestamps, and sizes.
`backup_purge`	destructive	Purge old backups, keeping only the most recent N. Useful for managing disk space.
`backup_restore`	read	Restore a previously created backup by ID. Destructive: replaces this user
`backup_verify`	read	Verify backup integrity using SHA-256 checksum. Use to check if a backup is corrupted before restoring.
`complete_todo`	read	Mark a todo as complete. For recurring tasks, automatically creates the next occurrence.
`consolidation_report`	read	Get a report of what the memory system has been learning. Shows memory strengthening/decay events, edge formation, fact extraction, and maintenance cycles. Use this to understand how your memories are evolving.
`context_summary`	read	Get a condensed summary of recent learnings, decisions, and context. Use this at the start of a session to quickly understand what you
`count`	read	Number of memories (default: 10)
`delete_project`	destructive	Permanently delete a project. Use delete_todos=true to also delete all todos in the project.
`delete_todo`	destructive	Delete a todo permanently.
`delete_todo_comment`	destructive	Delete a comment from a todo.
`dismiss_reminder`	read	Dismiss/acknowledge a triggered reminder. Call this after you
`fact_narratives`	read	Get synthesized narratives from accumulated facts, clustered by topic with confidence levels and causal chains. Shows what the system has learned, organized into coherent themes.
`forget`	destructive	Delete a specific memory by ID
`list_anomalies`	read	Rank recent memories by statistical deviation from this user
`list_documents`	write	List the documents in the corpus, with sizes. Start here: it tells you what is available
`list_entities`	read	List the entities in the knowledge graph, ranked by salience (learned importance), with their types and mention counts. Use to discover what people, systems, places, and concepts the graph tracks — or to find the exact name to pass to explore_entity. For memory contents use recall; this is the graph
`list_memories`	read	List all stored memories
`list_projects`	read	List all projects with todo counts and status breakdown.
`list_reminders`	read	List all pending reminders. Use to check what reminders are scheduled.
`list_subtasks`	write	List subtasks of a parent todo. Use add_todo with parent_id to create subtasks.
`list_todo_comments`	read	List all comments and activity history for a specific todo.
`list_todos`	read	List or search todos. Supports semantic search via query parameter, or GTD-style filtering. Returns Linear-style formatted output grouped by status.
`memory_health`	read	Check memory system status and statistics
`memory_stats`	read	Get statistics about stored memories
`pending_work`	read	Show todos and incomplete tasks
`proactive_context`	read	REQUIRED: Call this tool with EVERY user message to surface relevant memories and build conversation history. Pass the user
`purge_facts`	destructive	Delete facts matching a content pattern. Use dry_run=true to preview before deleting. Useful for cleaning up garbage facts (e.g.,
`query`	read	What to search for in memories
`quick_recall`	read	Search your memories for relevant context
`read_document`	read	Read one text document from the corpus. Returns the head of very large files with a note
`read_memory`	read	Read the FULL content of a specific memory by ID. Use this when you need to see the complete text of a memory that was truncated in search results.
`recall`	read	Search memories AND todos using semantic similarity. Returns both relevant memories and matching todos. Use this to find past experiences, decisions, context, or pending work. Modes:
`recall_by_tags`	read	Find memories by tags. Returns memories matching ANY of the provided tags. Useful for finding memories by category (e.g.,
`recent_memories`	read	Show recently created memories
`reinforce_memories`	read	Give Hebbian feedback on memories after using them: outcome
`remember`	read	Store a memory for future recall. Use this to remember important information, decisions, user preferences, project context, or anything you want to recall later.
`reorder_todo`	write	Move a todo up or down within its status group. Use to prioritize tasks manually.
`repair_index`	read	Repair vector index by re-indexing orphaned memories. Use this when verify_index shows unhealthy status. Returns count of repaired memories.
`reset_token_session`	destructive	Reset the token counter for a new session. Call this when starting a new conversation or after context has been compressed/summarized.
`search_documents`	read	Find which documents contain a string, with the matching lines. Use before read_document
`session_digest`	read	Get a consolidated digest of the current session: timestamps, token usage, memories created/recalled, tools used with counts, entities extracted, topic changes, and consolidation events. Use after context compression or at session milestones.
`session_history`	read	Show what you worked on across recent sessions. Returns session summaries with entities, memory stats, and timestamps. Use group_by_project to detect cross-session project continuity via entity overlap.
`session_summary`	read	Get a summary of recent learnings, decisions, and context
`set_reminder`	write	Set a reminder for the future. Triggers on time (at specific time or after duration) or context match (when keywords appear in conversation). Reminders will surface automatically when conditions are met.
`shodh_add_todo`	write	Create a task/todo with optional project, priority, contexts, and due date. Follows GTD methodology.
`shodh_complete_todo`	read	Mark a task as complete. For recurring tasks, automatically creates the next occurrence.
`shodh_context_summary`	read	Get a condensed summary of recent decisions, learnings, and project context from memory.
`shodh_forget`	destructive	Delete a specific memory by its ID. Use when information is outdated or incorrect.
`shodh_list_todos`	read	List tasks/todos filtered by status and project. Returns tasks with IDs, priorities, and status.
`shodh_proactive_context`	read	Surface memories relevant to the current conversation context. Uses entity matching and semantic similarity.
`shodh_recall`	read	Search persistent memory using natural language. Returns semantically similar memories ranked by relevance.
`shodh_remember`	read	Store a memory that persists across sessions. Use for important facts, decisions, user preferences, or learnings.
`todo_stats`	read	Get statistics about your todos - counts by status, overdue items, etc.
`token_status`	read	Get MCP pipeline token throughput for this session. Tracks tokens flowing through shodh memory tools only — NOT the AI context window. Use for internal diagnostics, not context window health.
`topic`	read	The topic to explore
`update_todo`	write	Update a todo
`update_todo_comment`	write	Update an existing comment on a todo.
`validate_causal_link`	read	Confirm or reject a causal edge by its edge ID (shown by trace_lineage and list_causal_edges). Confirming an inferred edge raises it to full confidence and strengthens the knowledge-graph connections between the two memories
`verify_index`	read	Verify vector index integrity - diagnose orphaned memories that are stored but not searchable. Returns health status and count of orphaned memories.
`what_i_know`	read	Surface everything related to a topic

Trust audit

BLOCKgrade F · trust 54/100 Do not install this without reading the findings. The audit found something that could harm you or your machine.

Layer	What it checks	Result
L0	Provenance & inventory	WARN
L1	Static analysis of the code	FAIL
L2	Instruction surface (what it tells the agent)	WARN
L3	Class-specific surface	WARN
L4	Behavioural (sandbox)	SKIPPED

What the source does

Filesystem: declared (11 observation(s))
Network: declared (5 observation(s))
Shell: declared (1 observation(s))
Dependencies: not all pinned
Secrets in source: found