Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/slicervm/agent-skillsnpx agentmods add skills/slicervm/agent-skills/use-dual-terminal-raceWrote 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/slicervm/agent-skills/use-dual-terminal-race)<a href="https://agentmods.dev/skills/slicervm/agent-skills/use-dual-terminal-race"><img src="https://agentmods.dev/badge/skills/slicervm/agent-skills/use-dual-terminal-race/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/slicervm/agent-skills/use-dual-terminal-race"><img src="https://agentmods.dev/badge/skills/slicervm/agent-skills/use-dual-terminal-race.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 | $0.00077 | $0.03097 |
| Opus 5 | $0.00039 | $0.01548 |
| Sonnet 5 | $0.00015 | $0.00619 |
| Haiku 4.5 | $0.00008 | $0.00310 |
Grade B, and why
use-dual-terminal-race scanned grade B with 2 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 12d 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.
Sends data to an external URLmediumData exfiltration
A POST to an outside endpoint may be telemetry or may be exfiltration; either way the mod talks to somewhere, and you should know where.
slicer vm exec ds -- 'curl -fsS -m 30 <endpoint>/v1/chat/completions -H "Content-Type: application/json" -H "Authorization: Bearer <key>" -d "{\"model\":\"<model>\",\"messages\":[{\"role\":\"user\",\"content\":\"Reply wi Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
slicer vm exec ds -- 'curl -fsS -m 30 <endpoint>/v1/chat/completions -H "Content-Type: application/json" -H "Authorization: Bearer <key>" -d "{\"model\":\"<model>\",\"messages\":[{\"role\":\"user\",\"content\":\"Reply wi How it starts
The opening of the file, as written. The whole thing — 259 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Use Dual-Terminal Race — two agents, two VMs, one video
Two opencode agents, each in its own fresh Slicer microVM, run the same task
side by side and are recorded as one video. The terminals are real xterms
on a real X server (Xvfb) on the host; each xterm bridges into its VM
with slicer vm shell and attaches to a guest tmux session.
This is the two-VM, host-display variant of use-xvfb-terminal-recording
(which runs Xvfb and xterm inside a single VM).
Why this shape
- One VM per agent: each model gets a clean, identical machine; a failure or a stuck agent on one side cannot corrupt the other.
- Xvfb + xterm on the host:
slicer vm shellgives a clean PTY per VM; xterm renders it with a real font; ffmpeg x11grabs the whole screen. No nested display plumbing, no per-VM X server. - Guest tmux (
demosession): the agent survives xterm/bridge reconnects, and you can drive and observe it withtmux send-keys/tmux capture-panewithout touching the display.
Prerequisites (host, once)
xvfb,xterm,fonts-dejavu-core,ffmpeg,imagemagick(import),x11-utils(xwininfo) on the host.- Slicer daemon running;
opencode+~/.local/share/opencode/auth.json+~/.config/opencode/opencode.jsonon the host (they get copied into the VMs). - An inlets tunnel for delivery (optional).
Start Xvfb if not running (must persist across execs):
DISPLAY=:99 Xvfb :99 -screen 0 1920x1080x24 -ac +extension GLX +render -noreset &
pgrep -a Xvfb && ls /tmp/.X11-unix/ # -> X99
Step 1 — Provision the two VMs
slicer opencode --name ds --tag workflow=k3sup-race --tag model=deepseek
slicer opencode --name qw --tag workflow=k3sup-race --tag model=qwen
- No path argument: provision-only (VM + git identity + opencode creds + config + cached opencode binary). Nothing is copied from the host.
--nameis essential: the allocated hostname (e2e-N) is unpredictable after deletions; the friendly name is the stable handle.slicer opencodecopies the host's opencode config and auth into each VM.
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.
- 12d ago First seen · 259 lines · 77 tokens per session scan B 597e65a6878e
use-dual-terminal-race is a skill published in the GitHub repository slicervm/agent-skills (5 stars, last pushed yesterday), licensed MIT. It adds 77 tokens to every session and 3,097 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it B with 2 findings (sends data to an external url, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…