MakersMCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| CHROME_PATH | No | Path to Chrome, Chromium, Edge, or Brave when a kit cannot find it automatically. | |
| FFMPEG_PATH | No | Path to ffmpeg when a kit cannot find it automatically. | |
| MAKERSMCP_OUT | No | The folder where each tool saves its files. Defaults to the directory where makersmcp starts, or ~/makersmcp when it cannot write there. | |
| MAKERSMCP_KITS | No | Turns on only the listed kits, such as easybrandkit,launch-film. Leave it out or set it to all to enable every kit. | all |
| LAUNCH_FILM_CHROME | No | Path to Chrome used by the launch-film kit. | |
| LAUNCH_FILM_FFMPEG | No | Path to ffmpeg used by the launch-film kit. | |
| MAKERSMCP_KIT_PATHS | No | Uses a kit's checkout instead of its package, specified as <kit>=<dir>. |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
MakersMCP runs open source kits on this machine: brand kits, fonts, screenshots, demo videos, launch films and more, each from a website or a file. Every tool saves its files in a folder (relative ones go under /app) and returns the folder, the files and the kit's own notes. A tool that is still working after wait_seconds returns a job_id: call wait_for_job with it, never the same tool again. check_setup tells what is missing when a tool fails to start.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| make_brand_kitA | Reads a website in a headless browser, the way it is drawn, and saves its brand in one folder: the logo files cut from their real background (PNG and SVG), the colors with where each one is used, the fonts, the first screen, a brand card, and the app store icon and screenshots when the site links to its apps. BRAND.md is for people and agents, brand.json for tools, _brand.yml for Quarto and Shiny. With facts, it also lists what the site claims about itself, each claim with its page, in FACTS.md. Read BRAND.md before designing anything for that product. (easybrandkit 0.4.0, on this machine.) |
| make_launch_filmA | Makes a 20 to 40 second launch film from a product's own site: it reads the brand and the page, plans the film (name and promise, the real screen in a browser window, the page's own headline, feature cards, a number the page states, the logo close) and renders film.mp4 with music and sound effects, all with Chrome and ffmpeg on this machine. Nothing is generated by AI: the film shows what the site says. film.json is the plan; edit it and call render_launch_film to render again. Rendering takes a few minutes. (launch-film 0.2.0, on this machine.) |
| plan_launch_filmA | Reads a product's brand and page and writes film.json, the plan of a launch film, without rendering it. Edit film.json (launch_film_guide with topic types lists the scene types and their fields), then call render_launch_film. (launch-film 0.2.0, on this machine.) |
| render_launch_filmA | Renders a film.json into film.mp4 next to it, with Chrome and ffmpeg on this machine. Use it after plan_launch_film or after editing the film.json of make_launch_film. Rendering takes a few minutes. (launch-film 0.2.0, on this machine.) |
| make_launch_stillsA | Makes the pictures a launch needs at the exact size each place asks for: Product Hunt (ph), link previews (og), X (x), Google Play (play) and the Chrome Web Store (chrome). They come from the film's look, the brand's logo and the site's real screen, as opaque PNGs in stills/, listed in STILLS.md. Give a site and it plans a film first, or give a film.json so the pictures match the film. (launch-film 0.2.0, on this machine.) |
| check_videoA | Checks any rendered video by measuring it: size, frame rate, length, black, still and silent stretches, loudness, peaks, and whether the cuts land on the beat. Writes CHECK.md (pass or fail, with every check), check.json, sheet.png (a frame from each shot) and wave.png. A failed check is a normal result here; read CHECK.md for what to fix. (launch-film 0.2.0, on this machine.) |
| launch_film_guideA | Lists launch-film's looks and what each one suits (topic looks), or its scene types and their fields (topic types), for picking a look or editing film.json. (launch-film 0.2.0, on this machine.) |
| wait_for_jobA | Waits for a job a makersmcp tool started and returns its result once it ends. If it is still running, call this again with the same job_id. Never start the original tool again to check on a job: that starts a second one. |
| cancel_jobA | Stops a running job and the browser or ffmpeg it started. Files it already saved stay where they are. |
| list_jobsA | Lists the jobs makersmcp started since it was launched, running or ended, with their folders. |
| check_setupA | Shows which kits makersmcp has and their versions, whether Chrome and ffmpeg were found, and where files are saved. Use it when a tool is missing or a kit fails to start. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 11 tools
Most tools have clear, distinct targets: job management (list/wait/cancel), setup inspection, brand kit creation, video checking, and film guide. The only real overlap is make_launch_film versus plan_launch_film + render_launch_film, but the descriptions explicitly clarify the one-shot versus decomposed workflows. An agent should be able to select correctly after reading the descriptions.
All names are snake_case and mostly follow a verb_noun pattern (list_jobs, make_brand_kit, plan_launch_film, render_launch_film, check_video, wait_for_job, cancel_job). The main deviation is launch_film_guide, which is noun-only rather than verb-led, but it is still readable and predictable.
Eleven tools is well within the ideal range and each tool earns its place: three job tools, one setup check, three brand/film planning tools, two film rendering/stills tools, one video checker, and one guide. The decomposition of plan/render is justified by the iterative editing workflow.
The surface covers the full launch-film lifecycle: brand extraction, planning, rendering, still generation, video quality checks, guidance, and async job control. Minor gaps remain, such as no explicit tool to delete or prune jobs and no tool to read back brand kit outputs other than reading files on disk, but these are workable.