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/openonion/connectonion/cli-skill-designnpx skills add openonion/connectonion --skill cli-skill-designgit clone --depth 1 https://github.com/openonion/connectonionWrote 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/openonion/connectonion/cli-skill-design)<a href="https://agentmods.dev/skills/openonion/connectonion/cli-skill-design"><img src="https://agentmods.dev/badge/skills/openonion/connectonion/cli-skill-design.svg" alt="Measured on agentmods" 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 | $0.00079 | $0.02061 |
| Opus 5 | $0.00039 | $0.01030 |
| Sonnet 5 | $0.00016 | $0.00412 |
| Haiku 4.5 | $0.00008 | $0.00206 |
Grade A, and why
cli-skill-design 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 4d 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 — 187 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Designing a CLI skill
A CLI skill is two files that have to agree: the command surface (co <thing> ...)
and the SKILL.md that tells an agent how to drive it. Design them together —
the agent's whole world is what it types and what comes back.
co-browser is the worked example. Read it before you write anything: its exit-code
table, its "read the output, not just the exit code" rule, and its Done checklist are
what this methodology generalizes. (Building a skill that drives a website through
co browser? That is the sibling skill browser-workflow-skill-builder — DOM,
selectors, verification scripts. This one is about the command surface itself.)
Two properties. Each has a test you run and paste the result of — not a principle you assert in the PR.
(a) Tip-tested discoverability
Rule: every command execution — success and failure — ends by naming the next command, spelled out, with the argument shape filled in.
Read one with: co gmail read <#> ✅ names the command
See the docs for more options ❌ names nothing
The tip test
A tip is good if an agent that has only that tip makes the right next call. That is testable, so test it:
from connectonion import llm_do
llm_do(
f"You just ran a shell command. Its full output was:\n\n{out}\n\n"
"Your goal: read the newest email. Reply with ONE shell command and nothing else.",
model="co/gemini-3.7-flash",
)
Use a text-only call (llm_do), not co ai. An agent with a shell will run the
command it picks — measured: the first attempt at this test executed co gmail read 1
and then co auth google against a real account. You are grading the reply, not the
mailbox.
- Pass — the reply is a command that exists and advances the goal (
co gmail read 1). - Fail — it invents a name (
co gmail open 1), asks for help, or replies with prose.
Rules for the harness, or the result means nothing:
- Give it the output only. No
SKILL.md, no--help, no conversation history — those are exactly the crutches the tip exists to replace. - Pin the model so a rerun compares like with like.
- Run it per command, not once. Score the whole surface in a table and paste it into the PR:
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.
- 4d ago First seen · 187 lines · 79 tokens per session scan A 8502e0f9f86c
cli-skill-design is a skill published in the GitHub repository openonion/connectonion (1,481 stars, last pushed yesterday), licensed Apache-2.0. It adds 79 tokens to every session and 2,061 once invoked, about $0.0004 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
edgeone skill scanner
Scan any agent skill for security risks before you install or use it. Powered by Tencent Zhuque Lab A.I.G (AI-Infra-Guard). 100% local static analysis — no file contents or credentials leave your device. Compatible with CodeBuddy, Cursor, Windsurf, Claude Code, OpenClaw and more. Triggers on: 这个 skill 安全吗, skill 安全扫描…
baby-sit
Monitor a GitHub pull request until CI is green, diagnose failures, and rerun only evidence-backed flaky GitHub Actions jobs.
nano-banana-pro-openrouter
Deterministic OpenRouter image generation adapter for Nano Banana Pro / Gemini image models. Use as skillexec when a meta-skill needs local image files and structured IMAGEREADY records without spawning an LLM agent.
skill-creator-linter
Internal tool (not user-invocable). Called by meta-skill-creator as a DAG step (kind: agent) to lint a candidate meta-skill SKILL.md against G1 (parse + reference check + xmlescape grep + structural lint) and G2 (scheduler dry-run with stub executors). Deterministic, sub-second, no LLM. Returns JSON diagnostics.
paper-abstract-author
Write the abstract after the paper body has been revised, using the final claims and evidence.
langbot-deploy
Deploy and configure a LangBot instance — Docker / Docker Compose, Kubernetes, the config.yaml model, the Box sandbox runtime, the plugin runtime, and the global API key. Use when installing, deploying, upgrading, or configuring LangBot in production or self-hosted environments. Triggers on "deploy langbot", "langbot…