MFLUX is a native MLX implementation of generative image models that runs locally on Mac computers. It is for generating images with supported models through command-line tools or a Python API. The catalogue skills and instruction support workflows built around these image-generation models.
Borrowing it
Nothing to install: this file belongs to mflux-community/mflux. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/mflux-community/mflux/main/.cursor/skills/mflux-testing/SKILL.mdgit clone --depth 1 https://github.com/mflux-community/mfluxWrote 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/mflux-community/mflux/mflux-testing)<a href="https://agentmods.dev/skills/mflux-community/mflux/mflux-testing"><img src="https://agentmods.dev/badge/skills/mflux-community/mflux/mflux-testing/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/mflux-community/mflux/mflux-testing"><img src="https://agentmods.dev/badge/skills/mflux-community/mflux/mflux-testing.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00027 | $0.00784 |
| Opus 5.5 | $0.00011 | $0.00314 |
| Sonnet 5.5 | $0.00005 | $0.00157 |
| Haiku 4.5 | $0.00003 | $0.00078 |
Grade A, and why
mflux-testing 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 2mo 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 — 57 lines — stays where its author put it; the contents beside it link to each section on GitHub.
mflux testing
This repo uses pytest with image-producing tests. Always preserve outputs for inspection and never update reference images unless explicitly asked.
When to Use
- You need to run tests (fast/slow/full) or debug failing tests.
- There are image/golden mismatches and you need to report paths/output for review.
Instructions
- Prefer the justfile test recipes:
just test-fast(fast tests, no image generation)just test-slow(slow tests, image generation)just test(default selection, skips slow model tests)just test-all(all except high-memory tests; slow tests download model weights)
- Always keep
MFLUX_PRESERVE_TEST_OUTPUT=1on test runs (already built into the justfile test recipes). - If a change affects defaults, config resolution, metadata fields, or CLI behavior, add or update tests that cover the changed behavior directly instead of relying only on manual verification.
- If tests fail:
- Summarize the failing test names and the key assertion output.
- Point to any generated images/artifacts on disk for manual review.
- Do not regenerate/replace reference (“golden”) images unless the user explicitly requests it.
Updating golden images (new model or hardware refresh)
Golden tests compare generated PNGs to tests/resources/reference_*.png (typically 15% mismatch threshold).
When to update (only with explicit user approval):
- After validating the port on target hardware (CI Mac) via slow tests with
MFLUX_PRESERVE_TEST_OUTPUT=1 - After choosing a stable prompt/seed/settings via diffusers comparison and/or latent-injection confidence (
mflux-debugging) - When old references used a bad prompt (e.g. ambiguous subject) or wrong seed for mflux’s RNG
Workflow:
- Run slow test → inspect
tests/resources/output_*.pngvsreference_*.png - If output is correct but reference is stale, re-run generation with same test parameters and replace reference PNGs
- Commit test + reference images together with a clear message (e.g.
test(<model>): update golden images for local hardware)
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.
- 2mo ago First seen · 57 lines · 27 tokens per session scan A b0d174ab869c
mflux-testing is a skill published in the GitHub repository mflux-community/mflux (2,547 stars, last pushed today), licensed MIT. It adds 27 tokens to every session and 784 once invoked, about $0.0001 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
verify-omnigent
Drive Omnigent the way a user does and prove a behavior with recorded evidence, using an isolated server, runner, and mock model plus a feature map of every user entry point. Load before reproducing a user-facing bug, before claiming a fix works, or when reviewing whether a change covered every surface (session vs.…
resolve-author-fix
Find the root cause, implement a focused fix, and prove behavior with targeted tests and recordings.
reproduce
Reproduces a paper's result, a claim, an artifact or a previous run with the target and success criterion frozen first, the canonical code path run before any substitute, exact inputs, environment, seeds and commands captured, and a verdict from a fixed set, reproduced, reproduced within tolerance, partially…
resolve-repro-audit
Audit a recovered repro patch, assertions, and behavioral baseline before authoring or reviewing.
otto-impact-assessment
Check the final diff's consumers, focused tests, evidence, and gaps.
feature-implement
Implement the design with focused tests mapped to each covered CUJ.