
Security News
Re-Enabled GitHub Actions Expose Thousands of Repositories to Mini Shai-Hulud
Two compromised GitHub Actions were re-enabled with malicious tags intact, exposing thousands of downstream repositories to Mini Shai-Hulud.
@noraai/mcp-server
Advanced tools
MCP server for Nora — deploy, monitor, and operate self-hosted OpenClaw and Hermes agent fleets on Docker or Kubernetes.
MCP server for Nora, the self-hosted AI agent ops platform. Connect Claude Code, Claude Desktop, Cursor, or any Model Context Protocol client to your Nora control plane and operate your agent fleet in natural language: deploy runtimes, start/stop/restart them, tail fleet status, and pull metrics, events, and per-agent cost.
claude mcp add nora \
--env NORA_API_URL=https://nora.example.com \
--env NORA_API_KEY=nora_xxxxxxxx \
-- npx -y @noraai/mcp-server
The repository also ships a Claude Code plugin manifest under
claude-plugin/. Validate and load it from a Nora checkout
with:
claude plugin validate ./mcp-server/claude-plugin --strict
claude --plugin-dir ./mcp-server/claude-plugin
Or in any MCP client's JSON config:
{
"mcpServers": {
"nora": {
"command": "npx",
"args": ["-y", "@noraai/mcp-server"],
"env": {
"NORA_API_URL": "https://nora.example.com",
"NORA_API_KEY": "nora_xxxxxxxx"
}
}
}
}
Build the same stdio server as a non-root container:
docker build -t nora-mcp ./mcp-server
docker run --rm -i \
-e NORA_API_URL=https://nora.example.com \
-e NORA_API_KEY=nora_xxxxxxxx \
nora-mcp
MCP clients should keep stdin open and pass secrets through environment variables; the image does not persist credentials or other state.
Uses Nora workspace API keys (create one under Workspace → API Keys). Scopes apply unchanged:
agents:read + monitoring:read → read toolsagents:write → deploy/lifecycle toolsFallbacks: NORA_HOST/NORA_TOKEN env vars, then the Nora CLI's ~/.nora/config.json — so after nora login, nora mcp (or plain npx @noraai/mcp-server) just works.
Read: list_agents, get_agent, get_agent_stats, get_agent_versions, get_platform_metrics, get_fleet_status, list_monitoring_events, get_agent_metrics, get_agent_metrics_summary, get_agent_cost.
Write: deploy_agent, start_agent, stop_agent, restart_agent, redeploy_agent — and delete_agent, which is only registered when NORA_MCP_ALLOW_DESTRUCTIVE=true.
Tool output is the raw Nora REST JSON; the server is a pure API client and stores nothing.
Full guide: noradocs.solomontsao.com/guides/mcp-server
The Nora MCP connector runs locally and does not collect telemetry or persist credentials, tool inputs, or API responses. It sends each tool request and the configured API key directly to the Nora deployment identified by NORA_API_URL; no separate connector service or advertising network receives that data.
The connector retains no data after its process exits. Your Nora operator controls the storage and retention of account, agent, log, metric, and integration data in that deployment, and connected providers may apply their own policies when Nora invokes them. See the full Nora Privacy Policy or contact privacy@solomontsao.com.
If Nora is useful to you, a ⭐ on the GitHub repo helps other self-hosters find it.
Apache-2.0
FAQs
MCP server for Nora — deploy, monitor, and operate self-hosted OpenClaw and Hermes agent fleets on Docker or Kubernetes.
The npm package @noraai/mcp-server receives a total of 23 weekly downloads. As such, @noraai/mcp-server popularity was classified as not popular.
We found that @noraai/mcp-server demonstrated a healthy version release cadence and project activity because the last version was released less than a year ago. It has 1 open source maintainer collaborating on the project.

Security News
Two compromised GitHub Actions were re-enabled with malicious tags intact, exposing thousands of downstream repositories to Mini Shai-Hulud.

Research
/Security News
A malicious Firefox extension fetches its payload after installation to evade detection, steal Google session cookies, and automate account takeover.

Research
/Security News
The compromise affects MemTensor's MemOS, an open source memory framework for large language models (LLMs) and AI agents. Both npm package @memtensor/memos-cloud-openclaw-plugin and the PyPI package MemoryOS are compromised. They drop cross-platform Go binaries that exfiltrate developer secrets.