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 automateyournetwork/netclaw --skill percepxion-oobgit clone --depth 1 https://github.com/automateyournetwork/netclawWrote 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/automateyournetwork/netclaw/percepxion-oob)<a href="https://agentmods.dev/skills/automateyournetwork/netclaw/percepxion-oob"><img src="https://agentmods.dev/badge/skills/automateyournetwork/netclaw/percepxion-oob/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/automateyournetwork/netclaw/percepxion-oob"><img src="https://agentmods.dev/badge/skills/automateyournetwork/netclaw/percepxion-oob.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 4 findings, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Data Exfiltration · line 151 Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.Fix: Verify the destination URL is trusted and necessary. Remove or replace with documented APIs. Ensure no secrets, tokens, or PII are transmitted.
- medium Data Exfiltration · line 157 Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.Fix: Verify the destination URL is trusted and necessary. Remove or replace with documented APIs. Ensure no secrets, tokens, or PII are transmitted.
- medium Data Exfiltration · line 1088 Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.Fix: Verify the destination URL is trusted and necessary. Remove or replace with documented APIs. Ensure no secrets, tokens, or PII are transmitted.
- medium Data Exfiltration · line 151 Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.Fix: Verify the destination URL is trusted and necessary. Remove or replace with documented APIs. Ensure no secrets, tokens, or PII are transmitted.
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 | $0.00070 | $0.13726 |
| Opus 5 | $0.00035 | $0.06863 |
| Sonnet 5 | $0.00014 | $0.02745 |
| Haiku 4.5 | $0.00007 | $0.01373 |
Grade A, and why
percepxion-oob 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 — 1,123 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Percepxion OOB Skill
What Is Out-of-Band Management?
Out-of-Band (OOB) management is a dedicated secondary path to the console port of every network device in your infrastructure. When a switch, router, firewall, or server becomes unreachable via its production (in-band) network interface, OOB gives you serial console access through an independent control plane, often with a resilient cellular network WAN connection, so you can, Day-0 provision, diagnose, recover, or remediate even when the production network is completely dark.
OOB turns a midnight outage that would require a truck roll into a routine remote session.
Lantronix hardware in this stack:
| Device | Role | Ports | Cellular | Status |
|---|---|---|---|---|
| SLC9000 | Console server, current gen. Serial or USB console ports and Ethernet switch ports. Dual power, redundant management interfaces, Percepxion-native, OpenAPI 3.1. | 16-48 | Optional 5G | Announced June 1st, 2026 |
| SLC8000 | Console server, previous gen. Still widely deployed in enterprise and carrier networks. Full Percepxion support. | 8-48 | - | Shipping until December 31st, 2026 |
| EMG series | Compact console server for small closets or remote edge sites. | 4-8 | Optional 4G | Production |
Percepxion is the SaaS-native management platform that aggregates these devices into a single API surface. It handles device ZTP, authentication, session brokering, firmware lifecycle, configuration management, access logging, and multi-tenant operations across thousands of devices. The Percepxion MCP server exposes 37 tools against this API.
Key Terms: OOB Device vs. Managed Device
This skill operates on two distinct device types. Confusing them causes wrong tool calls and unwanted outcomes.
| Term | What it is | Examples | How you reference it |
|---|---|---|---|
| OOB device (also: console server) | The Lantronix hardware managed by Percepxion. Has serial ports that cable to managed devices. | SLC9000, SLC8000, EMG7500, EMG8500 | By device_id in most MCP tool calls |
| Managed device (also: attached device, target device) | The network device whose console port is physically cabled to a serial port on the OOB device. NOT managed by Percepxion directly. | Cisco switch, Juniper router, Palo Alto firewall, OOB-connected server | Via get_security_telemetry (full inventory: hostname, model, serial, IP, OS) or get_port_telemetry (single port). list_device_ports returns port state only, not managed-device identity. |
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 · 1,123 lines · 70 tokens per session scan A e7d85f2b78a9
percepxion-oob is a skill published in the GitHub repository automateyournetwork/netclaw (657 stars, last pushed 5d ago), licensed Apache-2.0. It adds 70 tokens to every session and 13,726 once invoked, about $0.0003 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 2026-09-03.
Other skills, from other repositories
nbox-serve
Run nbox as an MCP server (read-only by default; optional local stdio writes with --local-writes, or shared HTTP/OIDC writes with --allow-writes + nbox:write + [serve.vault]) so an agent host can query NetBox over JSON-RPC — stdio or HTTP+OIDC transport, the read tools (search/get/getinterface/history/…), the…
nbox
Query and modify NetBox (DCIM/IPAM) from the shell with the nbox CLI — look up devices, interfaces, IPs, prefixes, VLANs, sites, racks, circuits, VRFs, tenants, VMs, and clusters; run cross-object search; use IPAM helpers (next free IP/prefix, prefix utilization, cable-path traces); and safely write (reserve…
nbox-device-context
Pull the full physical context of a device or interface from NetBox with nbox — a device's interfaces/IPs/cables/VLANs/services, an interface's cable-path A↔Z trace, a MAC reverse-resolved to its carrying interface/device, plus rack and site context. Use when the user asks how a device is connected, what's on an…
nbox-ipam-read
Answer IP addressing questions from NetBox with nbox's IPAM read commands — look up an IP and its parent prefix/VLAN/scope, check prefix utilization and the prefix tree, preview the next free IP/prefix, and inspect VLANs, IP ranges, aggregates, VRFs, and route targets. Use when the user asks about addressing…
nbox-patch-writes
Update NetBox object fields using nbox's PATCH write commands — interface set description and device set status. Use when the user wants to change an interface description or a device's status in NetBox.
aaa-auth-config
This skill should be used when the user asks to configure or troubleshoot AAA (Authentication / Authorization / Accounting) on Huawei S / CE / USG series devices — including 802.1X, MAC authentication, Portal authentication, RADIUS, HWTACACS (TACACS+), local authentication, authentication profiles, free-rule, NAC…