Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add sandraschi/bumi-mcp --skill bumi-operatorgit clone --depth 1 https://github.com/sandraschi/bumi-mcpWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/sandraschi/bumi-mcp/bumi-operator)<a href="https://agentmods.dev/skills/sandraschi/bumi-mcp/bumi-operator"><img src="https://agentmods.dev/badge/skills/sandraschi/bumi-mcp/bumi-operator.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00033 | $0.00645 |
| Opus 5 | $0.00016 | $0.00322 |
| Sonnet 5 | $0.00007 | $0.00129 |
| Haiku 4.5 | $0.00003 | $0.00064 |
Grade A, and why
bumi-operator scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 52 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Bumi operator
Description: Control and monitor Noetix Bumi humanoid robots. Covers motor control, sensor reading, status monitoring, safety protocols, and fleet deployment across multiple robot instances.
Trigger Phrases
- "Check Bumi robot status"
- "Get Bumi hardware specs"
- "Show me the virtual twin for Bumi"
- "Deploy Bumi fleet configuration"
- "What SDKs are available for Bumi?"
- "Start Bumi safety check"
Tools
bumi(operation="info")— Get general server info and available operations.bumi(operation="specs")— Return Noetix Bumi hardware specifications: degrees of freedom, sensor suite, compute module, battery specs.bumi(operation="sdk_links")— List OSS SDKs and API references for Bumi development.bumi(operation="robot_status")— Health check whenBUMI_ROBOT_URLpoints at a bridge. Returns connectivity state.bumi(operation="virtual_twin")— Return composition map for Bumi digital twin. References components that must be rendered in resonite-mcp or robotics-mcp.bumi(operation="fleet_peers")— List connected fleet MCP instances and their roles.bumi_agentic_workflow(goal="...")— High-level goal planning when client supports sampling (SEP-1577).
Physical Safety (CRITICAL)
- Do NOT command robot motion directly. bumi-mcp is a spec/bridge layer, not a locomotion controller.
- Use
robot_statusas HTTP health check only whenBUMI_ROBOT_URLis configured. - All physical operations require a vendor-approved API gateway with human-in-the-loop gate.
- Safety checks: confirm
robot_statusreturns operational before any fleet orchestration that involves the physical Bumi unit.
Virtual Twin
Call virtual_twin for the composition map. Implement traversal in resonite-mcp / robotics-mcp, not by expecting locomotion APIs inside bumi-mcp. The twin includes: joint hierarchy, sensor positions, skin mesh references, and interaction volumes.
Fleet Deployment
bumi(operation="status")— Confirm server up.bumi(operation="fleet_peers")— Discover connected MCP peers.bumi(operation="virtual_twin")— Get the composition if visual rendering needed.- Delegate locomotion/physics to robotics-mcp, visual rendering to resonite-mcp.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 8d ago First seen · 52 lines · 33 tokens per session scan A 01959bacfaee
bumi-operator is a skill published in the GitHub repository sandraschi/bumi-mcp (7 stars, last pushed 2d ago), licensed MIT. It adds 33 tokens to every session and 645 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
unifi-protect
How to manage UniFi Protect cameras and NVR — view cameras, smart detections, Find Anything detection search, recordings, snapshots, lights, sensors, Known Faces, license plates, and the Alarm Manager. Use this skill when the user mentions UniFi cameras, security cameras, NVR, recordings, motion detection, person…
unifi-access
How to manage UniFi Access door control — locks, credentials, visitors, access policies, and events. Use this skill when the user mentions UniFi Access, door locks, door access, building access, NFC cards, PIN codes, visitor passes, access policies, access schedules, door readers, or any UniFi Access task.
network-health-check
Run a UniFi network health check — diagnose device status, connectivity issues, firmware updates, and system health. Use when asked to check network health, find what's down, diagnose connectivity issues, or get a network status summary.
unifi-network
How to manage UniFi network infrastructure — devices, clients, firewall, VPN, routing, WLANs, Traffic Flows, and statistics. Use this skill when the user mentions UniFi, Ubiquiti, network management, WiFi configuration, firewall rules, port forwarding, VPN, QoS, bandwidth, traffic flows, connected clients, network…
unifi-access-support
Generate a sanitized UniFi Access support bundle, help the user review it locally, and prepare it for a bug report without posting it.
eplan-development
Develop EPLAN Electric P8 scripts, API extensions, and remote-control applications. Use when writing C# scripts for EPLAN (actions, event handlers, ribbon), accessing the EPLAN API (parts database, projects, pages), building external apps that drive EPLAN via Remote Client, or debugging EPLAN automation issues…