Borrowing it
Nothing to install: this file belongs to qorm/platform. 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/qorm/platform/main/AGENTS.mdgit clone --depth 1 https://github.com/qorm/platformWrote 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/instructions/qorm/platform/agents-md)<a href="https://agentmods.dev/instructions/qorm/platform/agents-md"><img src="https://agentmods.dev/badge/instructions/qorm/platform/agents-md/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/instructions/qorm/platform/agents-md"><img src="https://agentmods.dev/badge/instructions/qorm/platform/agents-md.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 | $0.03394 | $0.03394 |
| Opus 5.5 | $0.01358 | $0.01358 |
| Sonnet 5.5 | $0.00679 | $0.00679 |
| Haiku 4.5 | $0.00339 | $0.00339 |
Grade A, and why
platform AGENTS.md 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 — 196 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — orienting an AI agent in the QORM repo
What this is. QORM (Query · Observe · Render · Mutate) is a pure-Go,
agent-native cross-platform app platform. A QORM app is JSON — a manifest
(qorm.json) + scenes/*.json + actions/*.json, an optional Go
middle-layer, and an optional qorm.config.json (host window config:
size/title/resizable/chromeless/transparent — never bundled or signed) — that
one runtime renders to HTML/CSS, ed25519-signs into a
verifiable bundle, delivers over-the-air, packages for web / iOS / Android /
desktop, and exposes to agents over MCP. Dual-consumer by design: every artifact
is meant to be read, written, and verified by both a person and an AI. The
acronym is the API surface: Query (HTTP/MCP reads),
Observe (SSE), Render (the runtime), Mutate (actions + writes).
Full machine-readable map: llms.txt.
Understand it
- README.md — what QORM is + the CLI at a glance.
examples/— the canonical runnable apps. Trust these over any spec.
Write / edit an app
- One-Command Scaffold & Run: Run
qorm run <app-dir>(orgo run ./cmd/qorm run <app-dir>) — if<app-dir>doesn't exist, it automatically scaffolds the starter app and launches the native standalone executable application window for the platform. - Use the format the runtime accepts today: text via
text, bind with{{state.x}},onPressnames an action inactions/, components inqorm.jsonuse{{prop.x}}. See getting-started and the widget catalog (auto-generated from the code, canonical). - Do not use the old
value/on:{press}/{{count}}/scene://forms — the runtime ignores them. When docs and a runnable example disagree, the example wins. - App logic too big for steps goes in a script action: an action JSON with a
scriptfield holding qscript source (internal/qscript—let/if/for in/while/fn, the expression language's operators and builtins,stateread/write,argsin), or — the file-per-action spelling — anactions/<id>.qsfile whose filename is the action id and whose full text is the script. The loader compiles it (parse errors name the line, and the file for.qs);scriptbeatsstepswhen both are declared. Canonical example: examples/tetris. - Shared styles live in a stylesheet:
styles/<id>.qss(QORM Style Sheet,internal/qss) — rules likebutton { borderRadius: 12 },.accent { … },#submit { … }(type / class / id selectors;#comments; numbers stay numbers, strings and{{bindings}}evaluate like inline style values; nested objects such asmargin: {top: …}stay inline on the node). Scene nodes reference them with aclassprop (space-separated, later classes win). Cascade: theme component default < type rule < class rule (declaration order) < id rule < inlinestyle. Parse errors are load-time diagnostics naming file and line; unknown style keys warn. Canvas and HTML both apply matching rules (HTML merges them into emitted node CSS). Canvas visual effects (filter, mask, scroll-snap, FLIP, spring, text stroke/shadow, …) are documented in docs/styles.md and listed under common style props; showcase examples/canvas-fx. Example stylesheet: examples/tetrisstyles/app.qss. - No emoji in UI, code, or docs — use the built-in SVG icon set (icon names
like
heart/star/zap, listed ininternal/render/icons.go). - Style against the theme variables (
var(--accent),var(--label), …) so apps follow the OS light/dark setting (default theme isauto). ManifestdesignTokensrender as stage-scoped CSS vars too —color.primary→var(--qorm-token-color-primary). Unknownstylekeys are load-time warnings (the renderer ignores them).
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 · 196 lines · 3,394 tokens per session scan A 6a8c5971a820
platform AGENTS.md is an instructions file published in the GitHub repository qorm/platform (40 stars, last pushed 2mo ago), licensed MIT. It adds 3,394 tokens to every session, about $0.0136 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.
platform AGENTS.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| platform AGENTS.mdqorm/platform | Instructions file | 40 | 2mo ago | MIT | A | 3,394 | Codex, OpenCode |
| vscode buildNext.instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 9,719 | GitHub Copilot |
| vscode oss-third-party-notices.instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 5,516 | GitHub Copilot |
| next.js AGENTS.mdvercel/next.js | Instructions file | 143k | today | MIT | A | 7,461 | Codex, OpenCode |
| langchain AGENTS.mdlangchain-ai/langchain | Instructions file | 149k | today | MIT | A | 4,680 | Codex, OpenCode |
| vscode copilot-instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 3,093 | GitHub Copilot |
| vscode model-prompts.instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 2,022 | GitHub Copilot |