cli-and-defaults

Rules for using the Modellix command-line tool to run image, video, and audio tasks, including default models when none is chosen.

In plain words
What is it for?
It helps run Modellix generation, editing, speech, and transcription workflows through the CLI, with checks before paid work.
Why use it?
It gives the agent a consistent way to submit, wait for, and download tasks without writing its own polling code or choosing a model by guesswork.

Cursor rule

Install

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.

agentmods
npx agentmods add rules/modellix/modellix-plugin/cli-and-defaults
Clone the repo
git clone --depth 1 https://github.com/Modellix/modellix-plugin
Per session 352 This file is loaded in full into every session.
When invoked 352 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00352 $0.00352
Opus 5 $0.00176 $0.00176
Sonnet 5 $0.00070 $0.00070
Haiku 4.5 $0.00035 $0.00035

Measured yesterday against content hash 594d5b435975, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

cli-and-defaults 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 yesterday.

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.

rules/cli-and-defaults.mdc · 20 lines

What it actually says

  • Prefer modellix-cli over hand-rolled REST polling when the CLI is available.
  • Before the first CLI command in a workflow, run skills/modellix/scripts/preflight.py --json; it resolves public npm latest before paid work and keeps an existing CLI on update failure. Never change CLI versions after submission starts.
  • Canonical flow: automatic preflight/doctormodel run --waittask download. Prefer --json.
  • Do not reinvent poll loops when CLI --wait / task wait exists. Use REST only if CLI is missing or unsuitable.
  • If the user did not name a model, use these defaults immediately (do not scan the catalog first):
    • T2I: google/nano-banana-2-lite
    • T2V: bytedance/seedance-2.0-mini-t2v
    • I2I: google/nano-banana-2-lite-edit
    • I2V: bytedance/seedance-2.0-fast-i2v
    • V2V: bytedance/seedance-2.0-fast-v2v
    • TTS: alibaba/qwen-audio-3.0-tts-flash
    • STT: openai/whisper-1
    • STS: alibaba/cosyvoice-clone
  • Full policy and examples: plugin skill skills/modellix/SKILL.md and references/cli-playbook.md.
Changes

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.

  1. yesterday First seen · 20 lines · 352 tokens per session scan A 594d5b435975

Subscribe to this mod's changes

cli-and-defaults is a cursor rule published in the GitHub repository Modellix/modellix-plugin (1 stars, last pushed 20d ago), licensed MIT. It adds 352 tokens to every session, about $0.0018 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 2026-08-31.