Atlas / MCP servers / agentic-box / Memora

MemoraBLOCK

Name: Memora
Author: agentic-box

mcp/agentic-box/memora-1

MCP server by agentic-box · agent memory · ai agent · claude · Audited 2026-09-30

Give your AI agents persistent, collective memory — with deduplicating absorb, supersession lineage, semantic search, and a graph UI. Speaks MCP.

Verdict

BLOCK

Grade

Trust score

68 /100

Exposed tools

48 27r · 13w · 8d

Transport

streamable-http

License

MIT

Stars

729

Overview

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

Memora

"You never truly know the value of a moment until it becomes a memory."

Give your AI agents persistent collective memory An MCP memory layer for agents: structured storage, semantic retrieval, graph relations, and source-backed cross-session context.

Absorb agent work into durable graph memory, then use memory_digest(topic) to retrieve relevant memories, TODOs/issues, related edges, and source IDs.

Features · Preview · Install · Usage · Config · Multi-DB · Containers · Live Graph · Cloud Graph · Chat · Semantic Search · Documents · LLM Dedup · Linking · Neovim

Features

Core Storage

💾 **

Read from source at commit 51d480142fbcOBSERVED · 2026-09-30

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

claude mcp add memora-graph --env CF_API_TOKEN=${CF_API_TOKEN} --env CLOUDFLARE_API_TOKEN=${CLOUDFLARE_API_TOKEN} --env MEMORA_HEALTH_TOKEN=${MEMORA_HEALTH_TOKEN} --env OPENAI_API_KEY=${OPENAI_API_KEY} -- npx -y [email protected]

claude-desktop

{
  "mcpServers": {
    "memora-graph": {
      "command": "npx",
      "args": [
        "-y",
        "[email protected]"
      ],
      "env": {
        "CF_API_TOKEN": "${CF_API_TOKEN}",
        "CLOUDFLARE_API_TOKEN": "${CLOUDFLARE_API_TOKEN}",
        "MEMORA_HEALTH_TOKEN": "${MEMORA_HEALTH_TOKEN}",
        "OPENAI_API_KEY": "${OPENAI_API_KEY}"
      }
    }
  }
}

Exposed tools (48)

27 read · 13 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
`create_memory`	write	Create a new memory in the knowledge base. Use when the user asks to save, create, add, or remember something.
`delete_memory`	destructive	Delete a memory by ID. Use when the user asks to remove or delete a specific memory.
`memory_absorb`	read	Intelligently absorb facts into memory with dedup and consolidation.
`memory_backfill_tags`	read	Re-tag existing memories with project-prefixed tags.
`memory_boost`	read	Boost a memory
`memory_clusters`	read	Detect clusters of related memories.
`memory_create`	write	Create a new memory entry.
`memory_create_batch`	write	Create multiple memories in one call.
`memory_create_issue`	write	Create a new issue/bug memory.
`memory_create_section`	write	Create a new section/subsection header memory.
`memory_create_todo`	write	Create a new TODO/task memory.
`memory_delete`	destructive	Delete a memory by id.
`memory_delete_batch`	destructive	Delete multiple memories by id.
`memory_delete_document`	destructive	Delete a stored document and all its fragments.
`memory_detect_supersessions`	write	Detect memories that supersede (update/replace) other memories.
`memory_digest`	read	Return a deterministic digest of memories related to a topic.
`memory_events_clear`	destructive	Mark events as consumed.
`memory_events_poll`	read	Poll for memory events (e.g., shared-cache notifications).
`memory_export`	write	Export all memories to JSON format for backup or transfer. Rate limited: 60s cooldown.
`memory_export_graph`	read	Export memories as interactive HTML knowledge graph.
`memory_find_duplicates`	read	Find potential duplicate memory pairs with optional LLM-powered comparison.
`memory_get`	read	Retrieve a single memory by id (full content by default).
`memory_get_document`	read	Retrieve a stored document and its fragments by document key.
`memory_hierarchy`	read	Return memories organised into a hierarchy derived from their metadata.
`memory_hybrid_search`	read	Perform a hybrid search combining keyword (FTS) and semantic (vector) search.
`memory_import`	write	Import memories from JSON format. Rate limited: 60s cooldown.
`memory_import_sweep`	destructive	Admin: finish or remove rows an interrupted import left behind.
`memory_insights`	read	Analyze stored memories and produce actionable insights.
`memory_link`	write	Create an explicit typed link between two memories.
`memory_list`	read	List memories, optionally filtering by substring query or metadata.
`memory_list_compact`	read	[Deprecated] List memories in compact format (id, preview, tags only).
`memory_merge`	destructive	Merge source memory into target, then delete source.
`memory_migrate_images`	read	Migrate existing base64 images to R2 storage.
`memory_rebuild_crossrefs`	read	Recompute cross-reference links for all memories. Rate limited: 300s cooldown.
`memory_rebuild_embeddings`	read	Recompute embeddings for all memories. Rate limited: 300s cooldown.
`memory_related`	read	Return cross-referenced memories for a given entry.
`memory_semantic_search`	read	Perform a semantic search using vector embeddings.
`memory_stats`	read	Get statistics and analytics about stored memories.
`memory_store_document`	read	Store a structured document as a root memory + searchable fragments.
`memory_tag_hierarchy`	read	Return stored tags organised as a namespace hierarchy.
`memory_tags`	read	Return the allowlisted tags.
`memory_unlink`	destructive	Remove a link between two memories.
`memory_update`	write	Update an existing memory.
`memory_upload_image`	write	Upload an image file directly to R2 storage.
`memory_validate_tags`	read	Validate stored tags against the allowlist and report invalid entries.
`memory_verify_integrity`	read	Read-only embedding integrity doctor with bounded offending ids.
`ping`	read	return
`update_memory`	write	Update an existing memory by ID. Use when the user asks to modify, edit, or change a specific memory.

Trust audit

BLOCKgrade D · trust 68/100 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)	PASS
L3	Class-specific surface	WARN
L4	Behavioural (sandbox)	SKIPPED

What the source does

Filesystem: declared (3 observation(s))
Network: declared (9 observation(s))
Shell: none-observed
Dependencies: not all pinned
Secrets in source: found

Findings (25)

HIGHFilesystem / path · fs.credential_store · CWE-22, CWE-59

memora/backends.py:77

"  - AWS credentials file: ~/.aws/credentials\n"

Why it matters. touches a credential store

HIGHFilesystem / path · fs.credential_store · CWE-22, CWE-59

memora/server.py:3038

_BLOCKED_PATTERNS = [".ssh", ".gnupg", ".aws", ".config/gcloud", "id_rsa", "id_ed25519", ".env"]

Why it matters. touches a credential store

MEDIUMInventory / provenance · inv.symlink · CWE-1104

.claude/skills/memora

.claude/skills/memora

Why it matters. link not followed

MEDIUMInventory / provenance · inv.symlink · CWE-1104

memora-graph/public/_graph_limit.mjs

memora-graph/public/_graph_limit.mjs

Why it matters. link not followed

MEDIUMHard-coded secrets · secret.generic · CWE-798, CWE-321

tests/test_api_v1.py:21

TOKEN = "memora-api-v1-test-token"

MEDIUMFilesystem / path · mcp.destructive_tools · CWE-22, CWE-59

delete_memory, memory_delete, memory_delete_batch, memory_delete_document, memory_events_clear, memory_import_sweep, memory_merge, memory_unlink

Why it matters. 8 tool(s) can delete or overwrite

Fix. prefer a read-only mode or scoped tokens; the page states the blast radius

LOWInventory / provenance · inv.symlink · CWE-1104

memora-graph/public/index.html

memora-graph/public/index.html

Why it matters. link not followed