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 mikeholownych/unifi-mcp --skill unifi-mesh-backhaulgit clone --depth 1 https://github.com/mikeholownych/unifi-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/mikeholownych/unifi-mcp/unifi-mesh-backhaul)<a href="https://agentmods.dev/skills/mikeholownych/unifi-mcp/unifi-mesh-backhaul"><img src="https://agentmods.dev/badge/skills/mikeholownych/unifi-mcp/unifi-mesh-backhaul/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/mikeholownych/unifi-mcp/unifi-mesh-backhaul"><img src="https://agentmods.dev/badge/skills/mikeholownych/unifi-mcp/unifi-mesh-backhaul.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00000 | $0.00592 |
| Opus 5.5 | $0.00000 | $0.00237 |
| Sonnet 5.5 | $0.00000 | $0.00118 |
| Haiku 4.5 | $0.00000 | $0.00059 |
Grade A, and why
unifi-mesh-backhaul 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 2mo 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.
What it actually says
UniFi Mesh Backhaul Health
Diagnose slow zones caused by weak wireless uplinks between mesh APs — the invisible half of WiFi problems that channel tuning can't fix.
Mental model
A wirelessly-uplinked AP re-broadcasts everything over its backhaul radio: client throughput ≤ backhaul capacity ÷ 2 (same radio serves both sides), and every extra hop compounds the loss. A "great signal" client on a mesh node with a poor uplink still gets terrible speed.
Workflow
- Identify the topology: for each AP,
get_device_details— inspect theuplinkblock:is_wireless(true = meshed), remote AP, signal/strength, number of hops.- Wired uplink → backhaul excluded from suspicion.
- Wireless + strong (>-60 dBm equivalent) single hop → probably fine.
- Wireless + weak, or multi-hop (meshed onto another meshed AP) → prime suspect.
- Correlate complaints: map affected clients' APs (
get_client_details) against the weak-backhaul nodes. - Quantify: compare satisfaction/tx rates of clients ON the mesh node vs
clients on wired APs (
list_clients). - Fixes, in order of preference:
- Run Ethernet to the extender (wired backhaul = full speed; even cheap cable beats any mesh hop) — say this plainly, it's the real answer.
- Reposition the extender closer/more line-of-sight to its uplink AP.
- Reduce hops: re-home the node onto a wired AP rather than another mesh node.
- Last resort: remove the node if it does more harm than good.
- Verify post-change via step 1–3 again (uplink signal, client rates).
Output format
Mesh map:
[AP name] --(wireless, -67, 1 hop)--> [U6 LR] --(wired)--> [UDM Pro]
Suspect: [name/reason]
Options: [ranked fixes w/ tradeoffs]
Constraints
- Read-only skill: placement/cabling changes are physical actions for the user.
- Don't recommend more mesh nodes as a fix for mesh slowness (common trap).
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.
- 2mo ago First seen · 52 lines · 0 tokens per session scan A 68a61ec41cbd
unifi-mesh-backhaul is a skill published in the GitHub repository mikeholownych/unifi-mcp (2 stars, last pushed 13d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 592 tokens. 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.
unifi-mesh-backhaul compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| unifi-mesh-backhaulmikeholownych/unifi-mcp | Skill | 2 | 13d ago | MIT | A | 0 | Claude Code |
| gke-ai-troubleshooting-tpu-dynamic-slices-monitoringgoogle/skills | Skill | 22k | yesterday | Apache-2.0 | A | 107 | Claude Code, Codex |
| gke-ai-troubleshooting-tpu-vbar-oomgoogle/skills | Skill | 22k | yesterday | Apache-2.0 | A | 125 | Claude Code, Codex |
| doca-flowNVIDIA/skills | Skill | 3.9k | yesterday | Apache-2.0 | A | 224 | Claude Code, Codex |
| competition-firmware-layoutzhaoxuya520/reverse-skill | Skill | 44k | 1mo ago | MIT | A | 110 | Codex |
| diagnose-driver-installNVIDIA/deepops | Skill | 1.5k | yesterday | BSD-3-Clause | A | 47 | Claude Code, Codex |
| catc-troubleshootautomateyournetwork/netclaw | Skill | 715 | yesterday | Apache-2.0 | A | 114 | Claude Code |