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 sbrunner-atx/n3fjp-mcp --skill contest-operatinggit clone --depth 1 https://github.com/sbrunner-atx/n3fjp-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/sbrunner-atx/n3fjp-mcp/contest-operating)<a href="https://agentmods.dev/skills/sbrunner-atx/n3fjp-mcp/contest-operating"><img src="https://agentmods.dev/badge/skills/sbrunner-atx/n3fjp-mcp/contest-operating/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/sbrunner-atx/n3fjp-mcp/contest-operating"><img src="https://agentmods.dev/badge/skills/sbrunner-atx/n3fjp-mcp/contest-operating.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 | $0.00112 | $0.01401 |
| Opus 5 | $0.00056 | $0.00700 |
| Sonnet 5 | $0.00022 | $0.00280 |
| Haiku 4.5 | $0.00011 | $0.00140 |
Grade A, and why
contest-operating 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 10d 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 — 120 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Contest operating: exchange state machine and N3FJP logging
The QSO state machine (Field Day example)
Exchange = class + section (e.g., 2A STX). Substitute your contest's
exchange as needed.
CQ: "CQ FD CQ FD de <MYCALL> <MYCALL> pse k\n"
↓ caller heard (their callsign decoded)
EXCHANGE: "<CALL> de <MYCALL> <MYCALL>\nPse copy 2A 2A STX STX\nde <MYCALL> BK\n"
↓ they QSL and give (or already gave) their class+section
TU: "QSL <their CLASS> <their SECTION> TU <CALL> de <MYCALL> GL FD sk sk\n"
↓
LOG: log to N3FJP (see below), then back to CQ
Rules that improved completion rate during live operation:
- Double everything important.
2A 2A STX STXcopies far better than2A STXthrough QSB and QRM. - Echo their exchange back in the TU (
QSL 1D AZ TU W7XYZ ...). It lets the other station verify you logged them correctly before you QRZ. - Never log without a confirmed callsign. Class and section can be inferred from a repeat; the callsign cannot.
- A "CQ" from a station heard mid-loop is not a caller — see QRM below.
Special cases (all encountered live)
Caller sends callsign only, no exchange. Normal. Some operators send
just <MYCALL> de <THEIRCALL> and wait. Send your exchange; their
class/section arrives in the next over.
They already sent their exchange with the first call. Also normal
(K6ABC de W7XYZ 1D AZ 1D AZ pse k). Send your exchange; they reply QSL;
go straight to TU.
Caller didn't copy you (repeat, no cpy, AGN). Resend once, in
abbreviated form: <CALL> de <MYCALL>\n2A STX 2A STX\nBK\n. If still no
response, QRZ once, then return to CQ.
Garbled/partial callsign. Send de <MYCALL> AGN AGN pse k\n, wait one
listen window. If the repeat is still unreadable, return to CQ. Budget at
most two overs on an undecodable station.
Two stations doubled (called simultaneously). Output looks like two callsigns interleaved with control garbage. Same treatment as garble: one AGN, then CQ. Usually one of them calls again alone.
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.
- 10d ago First seen · 120 lines · 112 tokens per session scan A cfca5d0ad772
contest-operating is a skill published in the GitHub repository sbrunner-atx/n3fjp-mcp (0 stars, last pushed yesterday), licensed MIT. It adds 112 tokens to every session and 1,401 once invoked, about $0.0006 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
fldigi-operating
Operating discipline for driving fldigi via fldigi-mcp: keying the transmitter, returning to receive instantly after an over, and polling the RX buffer so no answer is missed. Use whenever conducting live QSOs, running a CQ loop, or any operation that alternates transmit and receive (contest exchanges…
pin-and-lock
Entering digits on native (non-IME) PIN pads, and setting/clearing a secure lock screen required by AndroidKeyStore / Keystore-backed crypto flows. Use when a task needs to unlock a device, enter a PIN, or set/clear a device lock.
immybot
Every ImmyBot endpoint typed, plus a local SQLite mirror that answers the cross-tenant questions the web UI cannot. Trigger phrases: what failed in last night's maintenance window, which tenants are still on an old version of chrome, why didn't this computer get the deployment, what changed in the fleet since…
unifi-network
Use when the user asks to audit a UniFi Network gateway: list adopted devices and connected clients, review firewall policies, ACL rules, VLANs and WiFi broadcasts, see what changed in site config since the last check, spot hardware that just joined the network, or check switch port and PoE headroom. Wraps the local…
keenetic-fundamentals
A reference for managing Keenetic routers through KeeneticOS, also called NDMS, and its interfaces. It covers the RCI JSON API and the Keenetic MCP server, which let software inspect or change router settings.
tdx-assets
This skill should be used when the user asks about TeamDynamix assets, inventory, hardware, or equipment. Common queries include "find an asset", "look up a serial number", "asset tag", "search for a laptop", "register a new device", "create a computer asset", "find a workstation".