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 agentmods add skills/barry-ran/quickdesk/shell-runnernpx skills add barry-ran/QuickDesk --skill shell-runnergit clone --depth 1 https://github.com/barry-ran/QuickDeskWhat 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 | $0.00051 | $0.00358 |
| Opus 5 | $0.00026 | $0.00179 |
| Sonnet 5 | $0.00010 | $0.00072 |
| Haiku 4.5 | $0.00005 | $0.00036 |
Grade A, and why
shell-runner 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 2d 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.
What it actually says
shell-runner
Run shell commands on a remote host connected through QuickDesk. Commands execute via cmd /C on Windows or sh -c on Unix.
Tools
run_command
Execute a shell command on the remote host.
- command (string, required) — the shell command to execute
- working_dir (string, optional) — working directory for the command
- timeout_secs (integer, optional) — timeout in seconds (default: 60)
Returns:
{
"exit_code": 0,
"stdout": "...",
"stderr": "...",
"stdout_truncated": false,
"stderr_truncated": false
}
Output is truncated at 512 KB per stream if the command produces large output.
Workflow
- Call
run_commandwith the desired command string. - Check
exit_code— 0 means success, non-zero indicates failure. - If
exit_codeis non-zero, inspectstderrfor error details. - For long-running commands, increase
timeout_secsbeyond the 60-second default. - Use
working_dirto execute commands in a specific directory without changing global state.
What ships with it
2 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.
- 2d ago First seen · 48 lines · 51 tokens per session scan A b6b7ae91af8e
shell-runner is a skill published in the GitHub repository barry-ran/QuickDesk (423 stars, last pushed 16d ago), licensed MIT. It adds 51 tokens to every session and 358 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 2026-08-30.
Other skills, from other repositories
test-skill
A simple test skill used by internal/afk/skill unit tests.
huawei-cloud-ascend-command
Huawei Ascend NPU natural language management skill, supporting both local direct connection and SSH remote modes. Provides comprehensive npu-smi command capabilities including device queries, configuration management, firmware upgrade, vNPU virtualization, certificate management, and compute power testing (FLOPS).…
pawflow-developer
Complete development guide for contributing to PawFlow: architecture, patterns, conventions, and key subsystems.
luffy-arm-fullpower-on
Use when the user explicitly asks to turn on, enable, open, or arm luffy-arm full-power mode, including "turn on full power", "enable full-power", 打开 full power, 开启全功率模式. Check layered host status first and reuse an already-ON dedicated gate; gate state and effective ADMINUSER access are distinct. The agent never…
computer-basics
How this bot uses its pinned Orgo cloud computer (opt-in load).
codex-audit
Делегировать в Codex CLI (read-only) одну из пяти задач - глубокий аудит/ревью кода, состязательное (adversarial) ревью со скептической позицией ("сломай решение"), независимое второе мнение по подходу/диффу, сверка реализации с исходным планом (compliance), или диагностику когда основной агент застрял. Использовать…