artelCAUTION
Your fleet's smart notepad — one that learns. Shared memory your agents write to and get back at the right moment, plus a ledger showing what the work cost and which chores keep getting done by hand. Self-hosted, MIT.
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
[](https://github.com/NicolasPrimeau/artel/actions/workflows/ci.yml) [](LICENSE.md) [](https://glama.ai/mcp/servers/NicolasPrimeau/artel) [](https://smithery.ai/servers/nicolas-primeau/artel) [](https://artel.run)
Your fleet's smart notepad. One that learns.
One pad that you and every agent you run write into. Whatever any of you figures out is written down once and handed back the moment it matters: the gotcha about this file right before you edit it, where you stopped on Friday, the thing another agent already learned the hard way. Nothing to file, nothing to tag, nothing to look up. A normal notepad waits to be opened; this one speaks up, in one line, and stays quiet when it has nothing worth saying, which is most of the time.
It also doesn't just accumulate. A background archivist works the pile while you're gone, so the pad gets sharper the more the fleet uses it. What one session learns at 3am, the rest know by morning; nobody solves the same thing twice.
You run it on your own machine. None of it goes to anyone's cloud.
What that looks like
Nobody opened a file to find any of that, and nobody had to know who to ask.
Qui
09f85579257dOBSERVED · 2026-10-10Connect
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 artel --env ANTHROPIC_API_KEY=${ANTHROPIC_API_KEY} --env ARTEL_AGENT_KEY=${ARTEL_AGENT_KEY} --env ARTEL_API_KEY=${ARTEL_API_KEY} --env ARTEL_KEY=${ARTEL_KEY} -- uvx artel{
"mcpServers": {
"artel": {
"command": "uvx",
"args": [
"artel"
],
"env": {
"ANTHROPIC_API_KEY": "${ANTHROPIC_API_KEY}",
"ARTEL_AGENT_KEY": "${ARTEL_AGENT_KEY}",
"ARTEL_API_KEY": "${ARTEL_API_KEY}",
"ARTEL_KEY": "${ARTEL_KEY}"
}
}
}
}Trust audit
CAUTIONgrade B · trust 84/100 Install with care. The audit found things worth knowing before you trust its output.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | WARN |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (5 observation(s))
- Network
- declared (16 observation(s))
- Shell
- declared (2 observation(s))
- Dependencies
- pinned
- Secrets in source
- none-found
Findings (23)
"lighthouse": "beacon",
"Lighthouse": "Beacon",
ARTEL_URL = "http://192.168.183.2:8000"
url = f"http://127.0.0.1:{port}{path}""PUBLIC_URL": f"http://127.0.0.1:{port}",page.goto(f"http://127.0.0.1:{port}/ui/login").gitleaks.toml
.pre-commit-config.yaml
.anchors.lock
@pytest.mark.parametrize("bad", ["/etc/passwd", "../../../etc/passwd", "a/../../b"])A Claude Code session connects via an MCP plugin and sees the API as a set of tools (`memory_write`, `memory_search`, `task_claim`, `message_send`, and so on). An AutoGen script just calls `httpx.post
docs/artel.gif
docs/code_review.gif
docs/demo-2.gif
docs/incident_response.gif
docs/mesh_network5.gif
The `memory_write` MCP tool's description already says `entry_type="memory" is the default and right for almost everything`. Update the docstring/description to mention that `directive` is a valid typ
- One-line, promptless install. New `GET /plugin/install` serves a script that registers an agent, persists `ARTEL_URL` / `ARTEL_AGENT_ID` / `ARTEL_API_KEY` to a sourced env file, then installs the pl
- One-line onboarding: `curl http://<host>:8000/onboard | sh`
curl -fsSL http://<host>:8000/onboard | sh
:`curl -fsSL ${base}/onboard | sh`;curl -fsSL http://<host>:8000/onboard | sh
Gates applied: no_behavioural_pass.
09f85579257dfull audit observations/trust-audit/mcp-server/nicolasprimeau__artel.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 |
|---|---|---|---|---|---|
| 2026-10-10 | 09f85579257d | CAUTION | B | 84 | first audit |
Questions
What is the artel MCP server?
Your fleet's smart notepad — one that learns. Shared memory your agents write to and get back at the right moment, plus a ledger showing what the work cost and which chores keep getting done by hand. Self-hosted, MIT.
Is artel safe to connect to an agent?
With care. The audit graded it B (84/100) and found 23 things worth knowing before you trust this server, listed below with the exact line each was found on.
What credentials does artel need?
It reads ANTHROPIC_API_KEY, ARTEL_AGENT_KEY, ARTEL_API_KEY, ARTEL_KEY, ARTEL_REG_KEY, CLAUDE_CODE_OAUTH_TOKEN, MCP_AGENT_KEY, OPENAI_API_KEY and OPENROUTER_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 does artel run?
It speaks stdio and streamable-http, so it runs as a local process your client starts. It is published on PyPI as artel.
How current is this page?
The grade is for one exact copy of the source (09f85579257d), read on 2026-10-10. The repository is watched and re-audited when it changes.