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 magnus919/agent-skills --skill headscale-derpgit clone --depth 1 https://github.com/magnus919/agent-skillsWrote 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/magnus919/agent-skills/headscale-derp)<a href="https://agentmods.dev/skills/magnus919/agent-skills/headscale-derp"><img src="https://agentmods.dev/badge/skills/magnus919/agent-skills/headscale-derp/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/magnus919/agent-skills/headscale-derp"><img src="https://agentmods.dev/badge/skills/magnus919/agent-skills/headscale-derp.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, 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 MCP Rug Pull · line 106 Docker image references without a specific tag (:latest is implicit) or digest (@sha256:...) can be silently replaced by a malicious image.Fix: Pin the image: image:tag or image@sha256:abc123
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 | $0.00058 | $0.01512 |
| Opus 5 | $0.00029 | $0.00756 |
| Sonnet 5 | $0.00012 | $0.00302 |
| Haiku 4.5 | $0.00006 | $0.00151 |
Grade A, and why
headscale-derp 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 9d 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 — 150 lines — stays where its author put it; the contents beside it link to each section on GitHub.
headscale-derp Skill
Overview
DERP = Designated Encrypted Relay Protocol. DERP is Tailscale's TURN-like fallback mechanism used when direct peer-to-peer NAT traversal fails. Traffic through DERP is fully encrypted (WireGuard inside TLS), but it routes through a relay server rather than directly between peers, so it introduces additional latency.
Common scenarios where DERP is used:
- Symmetric NAT — Both peers behind symmetric NAT gateways
- Corporate firewalls — Restrictive egress policies that block UDP/STUN
- Double NAT — Carrier-grade NAT on both sides
- Blocked STUN — UDP port 3478 is filtered
DERP in Headscale
Headscale includes an embedded DERP server that is enabled by default in the config.yaml:
derp:
server:
enabled: true
region_id: 999
region_code: "headscale"
region_name: "Headscale Embedded DERP"
stun_listen_addr: "0.0.0.0:3478"
private_key_path: "/var/lib/headscale/derp_server_private.key"
The embedded server runs a STUN endpoint on port 3478 and relays on port 443 (or whatever port Headscale listens on). It's suitable for small tailnets (under ~50 nodes).
Key config options
| Option | Default | Description |
|---|---|---|
derp.server.enabled |
true |
Enable the embedded DERP server |
derp.server.region_id |
999 |
Numeric region identifier (must be unique across all regions) |
derp.urls |
[] |
Additional DERP map URLs (for standalone servers) |
derp.paths |
[] |
Local DERP map JSON file paths |
derp.auto_update |
true |
Automatically fetch Tailscale's default DERP map |
derp.stun_listen_addr |
0.0.0.0:3478 |
STUN listener address |
DERP Map
A DERP map is a JSON structure that defines relay regions and nodes. Example:
{
"Regions": {
"900": {
"RegionID": 900,
"RegionCode": "us-nyc",
"RegionName": "New York",
"Nodes": [
{
"Name": "900a",
"RegionID": 900,
"HostName": "derp-nyc.example.com",
"DERPPort": 443,
"STUNPort": 3478,
"STUNOnly": false
}
]
}
}
}
What ships with it
5 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 9d ago First seen · 150 lines · 58 tokens per session scan A 53a9fd235028
headscale-derp is a skill published in the GitHub repository magnus919/agent-skills (76 stars, last pushed yesterday), licensed MIT. It adds 58 tokens to every session and 1,512 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
routine
Manage CocoRoutine Snowflake task scheduling. Usage: $routine list, $routine schedule --cadence , $routine disable , $routine delete .
terraform-expert
Expert-level Terraform infrastructure as code, modules, state management, and production best practices. Use when the user mentions infrastructure as code, devops, or automation, or when the task involves Terraform Basics, Resource Management, Modules, or State Management.
floom
Create, run, and manage Floom AI workers — cloud automations that run on a schedule or a trigger and report back. Reach for this whenever the user wants something to happen repeatedly or in the background rather than a one-off script: a recurring task, "every day / every hour / every week", "schedule this", "run it on…
bridge-health
Quick health check on the bridge daemon (CEO machine + any exec mirrors). Surfaces heartbeat staleness, version drift, active sessions, and the Phase 1 to Phase 2 adoption gate metrics. Use when the dashboard feels stale, the sync-pill is red, the daemon may have crashed, or before a /push-updates that changes daemon…
control-plane-ops
Operar o control-plane local-first do agents-lab com board canônico, long-runs bounded, handoff/checkpoint, rollout/rollback e espelhos externos sem perder governança.
gandi
Gère les domaines, DNS et emails via l'API Gandi v5. Permet de lister les domaines, consulter/modifier les enregistrements DNS, gérer les boîtes mail et configurer les redirections.