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 swkstudios/unifi-fabric-mcp-server --skill device-client-troubleshootinggit clone --depth 1 https://github.com/swkstudios/unifi-fabric-mcp-serverWrote 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/swkstudios/unifi-fabric-mcp-server/device-client-troubleshooting)<a href="https://agentmods.dev/skills/swkstudios/unifi-fabric-mcp-server/device-client-troubleshooting"><img src="https://agentmods.dev/badge/skills/swkstudios/unifi-fabric-mcp-server/device-client-troubleshooting/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/swkstudios/unifi-fabric-mcp-server/device-client-troubleshooting"><img src="https://agentmods.dev/badge/skills/swkstudios/unifi-fabric-mcp-server/device-client-troubleshooting.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 | $0.00030 | $0.00772 |
| Opus 5.5 | $0.00012 | $0.00309 |
| Sonnet 5.5 | $0.00006 | $0.00154 |
| Haiku 4.5 | $0.00003 | $0.00077 |
Grade A, and why
device-client-troubleshooting 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 2d 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 — 37 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Device and client troubleshooting
Diagnose with reads before any disruption. Pass MAC addresses: the statistics tools take
a MAC or Classic _id, never the Integration id. In discovery mode (the default) run each
tool below through call_read_tool or call_write_tool with name and arguments; in
full mode call it directly. For anything else use search_tools and describe_tool.
- Call
get_devicewithhost,siteand the device MAC asdevice_id. - Call
get_device_port_statewithhost,site, the MAC asdevice_idand optional 1-basedport_idx; check link, error and drop counters, and PoE. - Call
get_clientwithhost,siteand the client MAC asclient_id, thenget_client_link_diagnosticswithhost,siteand the sameclient_id(orclient_idsfor several clients, never both). - When a loop or blocked forwarding is plausible, call
get_device_stp_statewithhost,siteand the MAC asdevice_id. - Correlate timestamps, link state, diagnostics and STP; prefer a non-disruptive fix.
restart_deviceis the last resort, never before steps 1-4. Show the operator the device and the impact and wait for approval; there is no per-call preview, and astatus: "dry_run"result means dry-run mode and no restart. Restart one device, never retry automatically (read back after a timeout), then repeat steps 1-4.
Tool contracts checked against code (paths under src/unifi_fabric/)
| Tool | Mode | Required arguments (optional) | Return shape and source |
|---|---|---|---|
get_device |
read | host, site, device_id |
device object verbatim; tools/device_mgmt.py::get_device; readOnlyHint: true |
get_device_port_state |
read | host, site, device_id (port_idx) |
_device_record, port_table, lldp_table (or _port) verbatim; tools/statistics.py::get_device_port_state; readOnlyHint: true |
get_client |
read | host, site, client_id |
client object verbatim; tools/clients.py::get_client; readOnlyHint: true |
get_client_link_diagnostics |
read | host, site (client_id or client_ids) |
{meta, data} narrowed to the clients; tools/statistics.py::get_client_link_diagnostics; readOnlyHint: true |
get_device_stp_state |
read | host, site, device_id |
{meta, _device, _source, _stp, _ports, _device_record}; tools/statistics.py::get_device_stp_state; readOnlyHint: true |
restart_device |
write | host, site, device_id (reboot_type) |
upstream body verbatim, or {status, dryRun, message, preview, _consequence}; tools/device_mgmt.py::restart_device; readOnlyHint: false |
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.
- 2d ago First seen · 37 lines · 30 tokens per session scan A 1cdab556cdf8
device-client-troubleshooting is a skill published in the GitHub repository swkstudios/unifi-fabric-mcp-server (6 stars, last pushed 4d ago), licensed MIT. It adds 30 tokens to every session and 772 once invoked, about $0.0001 per session on Opus 5.5. 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 2026-11-04.
device-client-troubleshooting compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| device-client-troubleshootingswkstudios/unifi-fabric-mcp-server | Skill | 6 | 4d ago | MIT | A | 30 | Claude Code |
| unifi-access-supportsirkirby/unifi-mcp | Skill | 1.0k | yesterday | MIT | A | 31 | Claude Code |
| ips-cleanupSchimmilab/ipsymcon-mcp-server | Skill | 0 | today | MIT | A | 56 | Claude Code |
| unifi-network-supportsirkirby/unifi-mcp | Skill | 1.0k | yesterday | MIT | A | 31 | Claude Code |
| unifi-protect-supportsirkirby/unifi-mcp | Skill | 1.0k | yesterday | MIT | A | 32 | Claude Code |
| interface-confignanxiaoyao/network-huawei-skills | Skill | 2 | 3mo ago | MIT | A | 124 | Claude Code, Codex |
| ospf-confignanxiaoyao/network-huawei-skills | Skill | 2 | 3mo ago | MIT | A | 136 | Claude Code, Codex |