Stress Test logo

Stress Test

Community
machina-exm
stress-test

Combat-test draft film assets before locking them. Use after a passport and references exist and before any scene generation.

Overview

Publishermachina-exm
Repositoryfilm-studio-skills
Skill namestress-test
Stars
156
Forks
25
Bundled files
Instructions only
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by machina-exm on GitHub. Read the source before you install it.

Installation

Install the Stress Test AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/machina-exm/film-studio-skills/tree/main/skills/stress-test
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/machina-exm/film-studio-skills.git /tmp/film-studio-skills
mkdir -p .claude/skills
cp -r /tmp/film-studio-skills/skills/stress-test .claude/skills/stress-test
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Stress Test in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Stress Test on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Stress Test is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Stress Test

Test a draft asset under the conditions that can break it before locking it for film generation. Use cheap static images for the entire test.

Start

  1. Read the ## AI film studio section in the project's CLAUDE.md or AGENTS.md.
  2. Use the configured image model as the destination for test prompts.
  3. If the section is missing, tell the user to run setup first.
  4. Continue if requested by asking which image model they use and how they submit prompts.
  5. Check for /docs/registry.md, /docs/breakdown.md, and the asset passport under /assets/characters/, /assets/locations/, or /assets/props/.
  6. If the studio tree is missing, offer to create the needed folder, but do not invent a passport or a lock result.
  7. Ask which one draft asset to test.

Ask one question at a time throughout.

Verify the draft

Find the exact registry row by tag and version. Require a draft passport with a canonical descriptor and reference files. If the descriptor or references are missing, stop the lock attempt and name what must be completed in asset-passport. Never shorten or rewrite the descriptor while constructing test prompts.

Read the registry and /docs/breakdown.md to collect:

  • every scene containing the asset;
  • the lighting used in those actual scenes;
  • every asset that shares the frame with it;
  • the relevant state variant for each scene.

Ask the user about one unresolved scene condition at a time. Do not substitute a base asset for a state variant.

Build the combat-test matrix

Create a matrix inside the asset's passport under ## Stress test. Cover different angles, shot sizes, actual-scene lighting, and shared-frame pairings. Include a two-shot beside every asset it shares frame with. Use more than one matrix row when a pairing appears under materially different scene lighting.

Use these columns:

markdown
| Test | Angle | Shot size | Scene lighting | Paired asset | Prompt | Result | Verdict |
|---|---|---|---|---|---|---|---|

Use pending, pass, or miss in the verdict column. Keep each row traceable to a scene when it uses scene lighting or a pairing.

Write the test prompts

For every row:

  1. Paste the asset's canonical descriptor verbatim.
  2. Point to its approved reference files without renaming them.
  3. State the row's angle and shot size positively.
  4. Reproduce the lighting of the actual scene.
  5. For a two-shot, paste the paired asset's canonical descriptor verbatim and point to its references.
  6. Describe a simple static composition that exposes identity consistency without adding unnecessary action.

Do not turn these tests into video generations. Do not add cinematic complexity that the row is not testing.

Run and record

Give the prompts to the user for their configured image workflow, or use the access method they provided. Save each static result beside the passport under the matching /assets/<type>/ folder with a versioned test filename. Record each output path in the row's Result cell. Review every output against the passport and references. For characters, check face, hair, age cues, build, wardrobe, distinctive details, and variant state. For locations, check layout, landmarks, materials, palette, light character, and grade. For props, check form, scale, materials, markings, wear, and distinctive details. For two-shots, check both identities rather than only the target asset.

Mark any identity drift or broken required detail as miss. Do not average a miss away because another output looks good.

Apply the gate

For a character, require ten passes out of ten attempts. A result below 10/10 is a failed lock. For a location or prop, present the completed matrix and require an explicit pass decision from the user.

On any miss, keep the registry status draft and choose one next action with the user:

  • revise the descriptor or references in asset-passport, then rerun the full affected test;
  • consciously move the affected scene to the next production block.

Never hide a failed row or mark a partial test as locked.

On a pass:

  1. Change the exact registry row from draft to locked.
  2. Change the passport status to locked.
  3. Record the lock decision and completed result paths in its stress-test section.
  4. Preserve the passport descriptor verbatim.

Finish

Report the score or decision, every miss, the registry status, and the paths to the test record. State the gate law: generation for a scene cannot start until every character, state variant, location, and prop in that scene is locked. If any asset remains draft, name the blocked scenes and stop before expensive video generation.

Frequently asked questions

What does the Stress Test AI skill do?

Combat-test draft film assets before locking them. Use after a passport and references exist and before any scene generation.

Why use Stress Test on TypingMind?

Because you install it once and use it with any model. Stress Test is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Stress Test in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/machina-exm/film-studio-skills/tree/main/skills/stress-test. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Stress Test?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Stress Test?

As many as you like. As long as a model supports skills, you can use Stress Test with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Stress Test AI skill free?

It is published on GitHub by machina-exm. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇