memora

MCP server · GitHub · ★ 746 · graded 2026-10-27Source linked

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

No outside reports on file

Memora centralizes agent memory, which is powerful but risky if you don't control what information agents can store, who can access it, and how updates are validated.

Based on public adoption and transparency signals as of Oct 27, 2026. Our opinion, not a code audit or security guarantee.

Was this verdict for Memora helpful?

What it can access

Local filesNo local folder access notedLOW
Command executionNo shell command runner notedLOW
Network reachConnects to outside service APINOTED
CredentialsNo API token requiredLOW
Package sourceGitHub (agentic-box/memora)PUBLIC

Inferred from this listing’s public package details and description. Check the project’s source before installing.

Install

No npm or PyPI package is listed. Install from source: github.com/agentic-box/memora ↗

All setup options ↓

Check this skill Watch it in my stack Compare Add to allowlist README badge View project

About Memora

Memora is an MCP connector that gives AI agents a shared memory system. It stores information with deduplication, tracks how memories change over time, lets you search by meaning (not just keywords), and shows memories as a visual graph.

Who it’s for

Teams using AI agents who need them to remember things across conversations and share knowledge with each other. Useful for agents that need long-term context or need to learn from past interactions.

Alternatives

LangChain Memory modules: Built-in memory tools for single-agent conversations (simpler, less shared).
Replicache: A sync engine for local-first data that multiple clients can share (different architecture).
Pinecone or Weaviate: Vector databases for semantic search, often used for agent memory (no built-in deduplication or lineage).

Setup

Check these steps against the project’s own instructions before running anything.

Follow the project's README at https://github.com/agentic-box/memora (opens external site) for installation and Claude Desktop configuration details.

Why this grade

CC — Moderate risk signals

Signal-based grade (3/7 points): well starred on GitHub, public source code. This reflects adoption and transparency, not a code review.

Popularity and upkeep

31/100

GitHub stars: 746
Listing last refreshed: 2026-10-29

These are public popularity signals, not a maintenance check. Look at the repository’s commit history and open issues.

Security reports

No reports found for this package in our research (last checked 2026-10-29).

Check a specific version →

Risks to watch for

Data storage: Memora stores all agent memories in a database or backend; ensure sensitive information isn't accidentally absorbed into shared memory.
Deduplication confusion: The system tries to merge similar memories automatically, which could combine different but similar facts incorrectly.
Semantic search quirks: Memory retrieval by meaning (not exact words) might return irrelevant or hallucinated-sounding results if the similarity algorithm makes mistakes.
Access control: Without clear permissions, any agent connected to Memora could potentially read or modify memories from other agents.
Lineage tracking: The system tracks how memories change, but if an agent corrupts or lies in a memory update, it becomes part of the permanent record.

How to lower the risk

Least-privilege permissions

Only connect agents that need shared memory; isolate agents with sensitive data.
Restrict which agents can create, edit, or delete memories.

Network isolation

Run the Memora backend on a private network if possible, not exposed to the internet.
Use authentication (API keys, tokens) between agents and the memory store.

Monitoring

Log all memory reads, writes, and updates to detect unauthorized access.
Audit memory lineage regularly to catch corrupted or false memories early.

Red flags to watch for

Agents storing unencrypted secrets, credentials, or PII in shared memory.
Frequent memory conflicts or reversions suggesting tampering or bugs.
Semantic search results that seem nonsensical or unrelated to the query.

Questions

How do I install Memora?

Memora does not publish a standalone npm or PyPI package. Clone or follow the setup instructions in its source repository at github.com/agentic-box/memora.

Are there security reports about Memora?

No reports found for this package in our research (last checked 2026-10-29). Memora currently holds Grade C (Moderate risk signals) based on public adoption and transparency signals. Key risk areas to watch include data storage, deduplication confusion, semantic search quirks.

Is Memora actively maintained?

Memora has 746 GitHub stars. Our directory data was last refreshed on 2026-10-29. Check recent commit activity, release tags, and open issues at github.com/agentic-box/memora before pinning it in production.

What are alternatives to Memora?

Notable alternatives to Memora include LangChain Memory modules (Built-in memory tools for single-agent conversations (simpler, less shared)), Replicache (A sync engine for local-first data that multiple clients can share (different architecture)), Pinecone or Weaviate (Vector databases for semantic search, often used for agent memory (no built-in deduplication or lineage)). You can also compare Memora side by side with other MCP servers in the Related tools section on this page.

Maintain or use Memora? Show its live grade in your GitHub README, docs site, or run a check from the terminal before you install.

GitHub README Markdown:

markdown

[![AgentArmorHQ Risk Grade for Memora](https://agentarmorhq.com/api/badge/agentic-box-memora)](https://agentarmorhq.com/skills/agentic-box-memora)

HTML embed:

html

<a href="https://agentarmorhq.com/skills/agentic-box-memora"><img src="https://agentarmorhq.com/api/badge/agentic-box-memora" alt="AgentArmorHQ Risk Grade for Memora" height="20" /></a>

Terminal check:

terminal

curl -fsSL https://agentarmorhq.com/api/cli | node --input-type=module - check "agentic-box-memora"

Share with your security lead

Need sign-off to use memora? Copy this note for Slack, a ticket or a pull request.

### AgentArmorHQ Security Review Memo: memora
- **Listing URL:** https://agentarmorhq.com/skills/agentic-box-memora
- **Type:** App connector
- **Risk Grade:** Grade C (Moderate risk signals)
- **Outside Security Advisories on File:** 0
- **Public Adoption:** 746 GitHub stars

#### Permission & Blast-Radius Signals
- **Local files:** No local folder access noted
- **Command execution:** No shell command runner noted
- **Network reach:** Connects to outside service API
- **Credentials:** No API token required
- **Package source:** GitHub (agentic-box/memora)

#### Recommended Hardened Setup
- **Source Repo:** https://github.com/agentic-box/memora
- **Controls:** Pin an exact version in `.mcp.json`, scope access to a single workspace folder, and use a least-privilege read-only token where supported.

_Based on public adoption and transparency signals as of Oct 27, 2026. Our opinion, not a code audit or security guarantee._

Key facts

Type: App connector
Publisher: agentic-box
Source code: github.com/agentic-box/memora
Est. context load: ~850 tokens / turn (~5 tool schemas, 0.43% of 200k context)
Popularity: 746 GitHub stars
Last updated: 2026-10-29
Last graded: 2026-10-27

Community

Discussions for Memora

Ask questions or share configuration tips. Governed by our Community Rules & Terms.

No posts match this filter yet.

Be the first to ask a question or share a working tip.

Was this listing for Memora helpful?

Check before you install, in Claude

Your assistant answers Block, Review or No known issues for any package before it installs it.

Get this check inside Claude