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 dmae97/omk --skill cli-anythinggit clone --depth 1 https://github.com/dmae97/omkWrote 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/dmae97/omk/cli-anything)<a href="https://agentmods.dev/skills/dmae97/omk/cli-anything"><img src="https://agentmods.dev/badge/skills/dmae97/omk/cli-anything/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/dmae97/omk/cli-anything"><img src="https://agentmods.dev/badge/skills/dmae97/omk/cli-anything.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 | $0.00118 | $0.00940 |
| Opus 5.5 | $0.00047 | $0.00376 |
| Sonnet 5.5 | $0.00024 | $0.00188 |
| Haiku 4.5 | $0.00012 | $0.00094 |
Grade A, and why
cli-anything 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 1mo 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 — 82 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLI-Anything
Turn GUI-only software into something an agent can drive, without a display or a mouse.
The rule that makes it work
Build the data → call the real software → verify the output.
The harness is an interface to the application, never a replacement for it. It generates the app's native project file, then invokes the actual binary to render.
This one rule decides whether the harness is useful or a liability:
- The application is a hard dependency. If it is not installed, fail with install instructions. Never fall back to a Python library that approximates the render.
- A fallback renderer produces plausible output that is wrong in ways nobody checks. That is worse than an error, because the agent reports success.
When to reach for this
Use it when the target has no adequate CLI and the task needs a real artifact: a render, an export, a format conversion, a scripted edit.
Skip it when the software already ships a CLI that covers the task. ffmpeg,
pandoc, and ImageMagick are the harness you would have written.
How to build one
1. Find the engine. GUI apps separate presentation from logic. Locate the core
library and any CLI the backend already ships — melt for Shotcut, sox for
Audacity, libreoffice --headless, blender --background --python. Those are the
building blocks, not things to reimplement.
2. Read the native format. The project file is the data layer: MLT XML, ODF ZIP,
SVG, .blend driven by a bpy script. Manipulate it directly.
3. Map GUI actions to calls. Every button and menu item is a function call. If the app has undo/redo it almost certainly uses a command pattern, and those commands are your CLI operations.
4. Choose the interaction model. A stateful REPL suits agents that keep context; a subcommand CLI suits one-shot scripting. Supporting both is usually right. Decide what persists between commands and where that state lives.
5. Give the agent introspection. info, list, and status matter more here than
in a human CLI: an agent cannot look at the screen to see what it just did. Fail loudly
with unambiguous messages so it can self-correct, and keep commands idempotent.
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.
- 1mo ago First seen · 82 lines · 118 tokens per session scan A e0262e2085e6
cli-anything is a skill published in the GitHub repository dmae97/omk (148 stars, last pushed today), licensed MIT. It adds 118 tokens to every session and 940 once invoked, about $0.0005 per session on Opus 5.5. 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
repo-hygiene
Use when the scheduled repo-hygiene workflow runs from GitHub Actions (or an operator dry-run) to scan the repository for small, certain docs/test/code hygiene issues and fix them as one batched branch.
test-failure-explorer
Import and query an existing Vitest 3.2 JSON report to explore test failures, file-level errors and original diagnostic evidence.
money
Skill "money" from Hmbown/Codewhale, covering money, when to use, setup, workflow and non-goals.
interview
Ask one useful structured question at a time only when material product/implementation choices are genuinely missing; remember answers and produce a brief/spec. Discoverable facts should be investigated instead of asked.
document
Skill "document" from Hmbown/Codewhale, covering document, when to use, non-goals and workflow.
research
Current external research using available web tools, primary sources where possible, links/citations, date awareness, synthesis, and unresolved uncertainty.