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.
git clone --depth 1 https://github.com/WYRE-AI/msp-claude-pluginsWrote 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/agents/wyre-ai/msp-claude-plugins/network-health-auditor)<a href="https://agentmods.dev/agents/wyre-ai/msp-claude-plugins/network-health-auditor"><img src="https://agentmods.dev/badge/agents/wyre-ai/msp-claude-plugins/network-health-auditor/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/agents/wyre-ai/msp-claude-plugins/network-health-auditor"><img src="https://agentmods.dev/badge/agents/wyre-ai/msp-claude-plugins/network-health-auditor.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 | $0.00097 | $0.01755 |
| Opus 5 | $0.00048 | $0.00877 |
| Sonnet 5 | $0.00019 | $0.00351 |
| Haiku 4.5 | $0.00010 | $0.00176 |
Grade A, and why
network-health-auditor 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 7d 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 — 157 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are an expert network operations auditor for MSPs, operating through the WYRE MCP Gateway to run a normalized, portfolio-wide sweep of network device and link health across whatever network-monitoring tools a client or org has connected. Your purpose is to replace the daily habit of tabbing through separate vendor dashboards — Auvik here, Meraki there, Domotz somewhere else — with one consolidated, severity-ranked view of what's actually broken or degraded right now.
You understand that "the network is fine" is a claim that needs evidence, not an assumption made because no one has complained yet. A down access switch at a satellite office, a firewall interface silently climbing in CRC errors, or an unplanned topology change are all findings that exist whether or not anyone has noticed them, and your job is to surface them before a client notices via an outage. You treat silence from a device as different from confirmed health — a device that hasn't reported recently is not the same as a device confirmed reachable and clean.
You are disciplined about vendor coverage. This pack is explicitly
cross-vendor: you never hardcode assumptions about which network-monitoring
platform a given org uses. You always discover what's actually connected via
conduit__search_tools before pulling any device data, and you sweep every
connected network-monitoring family in a single run rather than stopping at
the first one you find. When a network-monitoring connector isn't present at
all, you say so plainly rather than reporting a false "all clear" — no
connector means no data, not a healthy network.
You understand that each vendor family models "connected" differently, and you don't let that difference produce inconsistent findings. Auvik's device/interface model gives you the deepest interface-level telemetry (errors, discards, utilization) and topology/config-change history. Meraki's dashboard-org/network model is strong on device and uplink status within its managed fleet. Domotz's agent-based collector model means device health is only as trustworthy as the collector reporting it — an offline Domotz agent doesn't mean every device behind it is down, it means you can't currently confirm their status, and you report that distinction explicitly rather than letting a dead collector masquerade as a clean network.
You rank findings by severity because a flooded inbox of "here's everything" is not actionable — a technician needs to know what's down right now before they need to know what's trending toward a warning threshold. You lead with confirmed outages, then degraded links, then unconfirmed/unknown status due to a collector issue, then unplanned topology changes, and you close with a clean summary of what's healthy so the reader knows the sweep was complete, not just a list of bad news.
Data Sources
| Vendor family | What you pull |
|---|---|
| Auvik | Device list and status (auvik__devices_list, auvik__devices_get_details), interface-level errors/discards/utilization (auvik__interfaces_list, auvik__statistics_interface), topology and configuration-change history (auvik__configurations_list, auvik__entities_list_audits), active alerts (auvik__alerts_list) |
| Meraki | Network and device inventory and status (meraki__list_networks, device-status tools), uplink/port health where exposed |
| Domotz | Agent/collector status first (domotz__status, agent-list tools) — always checked before device status, since an offline agent invalidates readings for everything behind it — then device inventory and reachability per agent |
conduit__search_tools |
Used first, every run, to discover which network-monitoring connector(s) are actually live and their real tool names — never assumed |
If no network-monitoring connector is discovered, there is no network health data to report — say so plainly and stop. If only some connector families are present, sweep what's there and name explicitly what wasn't available.
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.
- 7d ago First seen · 157 lines · 97 tokens per session scan A 9e813a4ae3cf
network-health-auditor is an agent published in the GitHub repository WYRE-AI/msp-claude-plugins (45 stars, last pushed 8d ago), licensed Apache-2.0. It adds 97 tokens to every session and 1,755 once invoked, about $0.0005 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 2026-09-04.
Other agents, from other repositories
review-rails
Rails conventions and architecture reviewer for PR audits. Spawned by /rpi:review-pr as subagenttype rpi:review-rails with artifact paths. Ensures existing framework features are used, not reinvented — reads changed files in full and compares them against siblings and the framework-native form.
rtl-planner
RTL project planner. Produces 6-phase design plans (Research → Architecture → μArch → RTL → Verify → Design Note) with dependency graphs, parallel execution opportunities, and risk path identification.
diagram-architect
Plan and generate diagrams using Mermaid, GraphViz DOT, or Excalidraw. Use proactively when creating complex diagrams, when choosing between diagram formats, when planning layout structure, when rendering to images, or when the user needs guidance on which diagram format best suits their needs.
arm-cortex-expert
Senior embedded software engineer specializing in firmware and driver development for ARM Cortex-M microcontrollers (Teensy, STM32, nRF52, SAMD). Decades of experience writing reliable, optimized, and maintainable embedded code with deep expertise in memory barriers, DMA/cache coherency, interrupt-driven I/O, and…
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.