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 miikkij/aimeat-protocol --skill aimeat-app-buildergit clone --depth 1 https://github.com/miikkij/aimeat-protocolWrote 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/miikkij/aimeat-protocol/aimeat-app-builder)<a href="https://agentmods.dev/skills/miikkij/aimeat-protocol/aimeat-app-builder"><img src="https://agentmods.dev/badge/skills/miikkij/aimeat-protocol/aimeat-app-builder/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/miikkij/aimeat-protocol/aimeat-app-builder"><img src="https://agentmods.dev/badge/skills/miikkij/aimeat-protocol/aimeat-app-builder.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 | $0.00074 | $0.01650 |
| Opus 5 | $0.00037 | $0.00825 |
| Sonnet 5 | $0.00015 | $0.00330 |
| Haiku 4.5 | $0.00007 | $0.00165 |
Grade B, and why
aimeat-app-builder scanned grade B with 2 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 9d 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.
Sends data to an external URLmediumData exfiltration
A POST to an outside endpoint may be telemetry or may be exfiltration; either way the mod talks to somewhere, and you should know where.
curl -s -X PUT "<upload_url>" -H "Content-Type: text/html" --data-binary @app.html Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -s https://aimeat.io/v1/prompts/build-app How it starts
The opening of the file, as written. The whole thing — 116 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Canonical version lives on the node. This baked-in copy is a bootstrap seed. If the
aimeat_skill_gettool is available, fetchnode:aimeat-app-builderand follow THAT version — it is kept current (research flow, pitfall registry, new tooling). Both versions agree on the core rule: the live spec atGET /v1/prompts/build-appalways wins.
Building AIMEAT apps
You are running as an AIMEAT-boosted OpenHands: an AIMEAT agent identity is connected
to this instance over MCP. The MCP tools are named aimeat_* (e.g. aimeat_app_publish,
aimeat_app_list, aimeat_extension_install, aimeat_storage_upload). Use them to
publish directly to the node — you do not need the AIMEAT web UI.
An AIMEAT app is a single self-contained HTML file. The node hosts it and serves it
on its own subdomain (e.g. https://<name>.apps.aimeat.io/). It logs the user in, reads and
writes their data, and uses node-hosted UI/AI libraries — all from <script> tags pointing at
the node. There is no build step and no backend to write: the node is the backend.
The one rule that matters: fetch the canonical spec first
The node serves the authoritative, always-current build-app specification. It is the
single source of truth for available libraries, allowed script URLs, the required <meta>
tags, auth/data APIs, and the templates. Before writing any app, fetch it and follow it
exactly:
curl -s https://aimeat.io/v1/prompts/build-app
Also pull the ready-made templates and library list (start from a template — do not invent structure):
curl -s https://aimeat.io/v1/app-templates # starter templates
curl -s https://aimeat.io/llms.txt # discovery: build prompt + template links
If the node base URL is not aimeat.io, use the one configured for this instance (see the
AIMEAT_BASE_URL env / the MCP server URL). Everything the app loads at runtime — CSS, auth,
data, UI libraries — must be a URL listed in that spec. Never invent script/style src
URLs (e.g. /i18n/translations.js, /components/*.js); those 404 and break the app. Only use
the libraries the spec names.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 9d ago First seen · 116 lines · 74 tokens per session scan B 3ec1975a12a9
aimeat-app-builder is a skill published in the GitHub repository miikkij/aimeat-protocol (7 stars, last pushed today), licensed MIT. It adds 74 tokens to every session and 1,650 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it B with 2 findings (sends data to an external url, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
grafloki
Start Grafana and Loki for centralized log collection from Kurtosis enclaves. View aggregated service logs in a Grafana dashboard. Use when you need a UI for browsing logs across multiple services or want persistent log storage.
architecture-paradigm-client-server
Applies client-server architecture for web/mobile apps. Use when designing systems with centralized backend services, trust boundaries, or offline-first sync.
panels
Give a repository an operator panel: a small web UI (dev server) the user opens from the sidebar to run and preview that repo, by adding an operator/ directory that is a runnable web app. Use whenever you scaffold or work in a repo the user should be able to open, preview, or operate from the sidebar.
frontend-design-taste
Build distinctive frontends that avoid generic AI aesthetics. Provides anti-slop enforcement, variance dials, art direction archetypes, creative arsenal, and performance guardrails. Use when creating new landing pages, UI components, design systems, or when the user says "make it look good", "design direction", "avoid…
mintlify-mdx
Mastering Mintlify's MDX pipeline — undocumented constraints, snippet sandbox rules, CSS override patterns, and npm bundling via yoink. Use when writing or debugging Mintlify snippets, customizing docs.json/styles.css, bundling npm packages into snippets, hitting cryptic MDX compile errors, or when the user says…
app-onboarding-questionnaire
Design and build high-converting questionnaire-style app onboarding flows modelled on proven conversion patterns from top subscription apps (Headspace, Noom, Duolingo). Use when building onboarding flows, sign-up funnels, questionnaire UIs, or conversion-optimized first-run experiences. Triggers on "onboarding flow"…