BeeSAFE
BeeMCP: an unofficial Model Context Protocol (MCP) server that connects your Bee wearable lifelogger to AI via the Model Context Protocol
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
[](https://badge.fury.io/py/beemcp) [](https://opensource.org/licenses/MIT)
Unofficial Model Context Protocol (MCP) server for interacting with your Bee wearable lifelogging data. More context on my blog.
This server acts as a bridge, allowing Large Language Models (LLMs) like Claude or custom AI agents to access and interact with your personal data stored in Bee, including conversations, facts, to-dos, and location history.
Disclaimer: This is an unofficial project and is not affiliated with Bee. Use it at your own risk. Ensure you understand the security implications of granting AI access to your personal data via the API key.
Core Concept
Bee.computer helps you capture moments from your life (conversations, places visited, notes). beemcp makes this data available to your AI assistant through the Model Context Protocol. This means you can ask your AI questions like:
- "What important things did I discuss last week?"
- "Remind me about Brad's dietary preferences."
- "Where was I last Tuesday afternoon?"
- "Add 'Book flight tickets' to my reminders."
The AI, using beemcp, can securely fetch or modify this information from your Bee.computer account.
Prerequisites
- Python: Version 3.10 or higher.
- Bee.computer API Key: You need an API key from your Bee account settings.
Installation
You can install and run beemcp using uv (recommended) or pip.
Using uv (recommended)
uv is a fast Python package installer and resolver. If you have uv installed, you don't need to install beemcp separately. You can run it directly us
0e4c2bd85702OBSERVED · 2026-10-08Connect
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 beemcp --env BEE_API_TOKEN=${BEE_API_TOKEN} -- uvx beemcp{
"mcpServers": {
"beemcp": {
"command": "uvx",
"args": [
"beemcp"
],
"env": {
"BEE_API_TOKEN": "${BEE_API_TOKEN}"
}
}
}
}Exposed tools (17)
11 read · 4 write · 2 destructive. Blast radius: 2 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 |
|---|---|---|
confirm-user-fact | write | # Update the fact with new confirmation status |
create-todo | write | new_todo = bee.create_todo(text, alarm_at) |
delete-todo | destructive | return |
delete-user-fact | destructive | return |
get-conversation | read | |
get-locations-by-time | read | |
get-locations-month | read | |
get-locations-today | read | |
get-locations-week | read | |
get-user-fact | read | |
list-all-conversations | read | |
list-all-locations | read | |
list-all-user-facts | read | |
mark-todo-completed | read | updated_todo = bee.update_todo(id, completed=True) |
record-user-fact | read | return |
update-todo | write | updated_todo = bee.update_todo(id, text, completed, alarm_at) |
update-user-fact | write | return |
Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | WARN |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (1 observation(s))
- Network
- declared (5 observation(s))
- Shell
- none-observed
- Dependencies
- pinned
- Secrets in source
- none-found
Findings (3)
delete-todo, delete-user-fact
* **Description:** Delete an existing todo. Only call this if the user explicitly says to delete a todo.
**Disclaimer:** This is an unofficial project and is not affiliated with Bee. Use it at your own risk. Ensure you understand the security implications of granting AI access to your personal data via t
Gates applied: no_behavioural_pass.
0e4c2bd85702full audit observations/trust-audit/mcp-server/okgodoit__bee.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 |
|---|---|---|---|---|---|
| 2026-10-08 | 0e4c2bd85702 | SAFE | B | 89 | first audit |
Questions
What is the Bee MCP server?
BeeMCP: an unofficial Model Context Protocol (MCP) server that connects your Bee wearable lifelogger to AI via the Model Context Protocol
What tools does Bee expose?
17 in total: 11 read-only, 4 that write, and 2 that can delete or overwrite (delete-todo, delete-user-fact). Every one is listed on this page with its risk.
Is Bee safe to connect to an agent?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean server reads B. Separately from the audit: 2 of its tools can destroy data, so scope the token you give it to what you actually need.
What credentials does Bee need?
It reads BEE_API_TOKEN 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 (0e4c2bd85702), read on 2026-10-08. The repository is watched and re-audited when it changes.