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 agentmods add instructions/microsoft/vscode/design-philosophygit clone --depth 1 https://github.com/microsoft/vscodeWhat 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 | $0.00577 | $0.00577 |
| Opus 5 | $0.00289 | $0.00289 |
| Sonnet 5 | $0.00115 | $0.00115 |
| Haiku 4.5 | $0.00058 | $0.00058 |
Grade A, and why
vscode design-philosophy.instructions.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 2d 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.
What it actually says
Design philosophy
When you touch a visual surface, reason about it in design terms, not pixels. Design-system spacing, radii, type roles, and colors use named tokens/ramps where available. Raw literals remain valid for structural measurements or when no token exists.
Work in three layers - name the feeling, find the principle it breaks, then reach for the move that restores it:
- Values (how it should feel): Calm · Focused · Consistent · Delightful
- Principles (questions to ask):
- Quiet at rest, present on intent - is anything shouting when nothing's happening?
- Room to breathe - does this feel cramped, tight, or jagged?
- The interface explains itself - could someone understand this at a glance?
- One thing leads; the rest supports - where does the eye go first?
- Elevation is encoded, not eyeballed - does roundness match how far the surface floats?
- Sameness signals sameness - do two like things look identical, and is any difference intentional?
- Delight earns its keep - what does this help the user do?
- Moves (the mechanics): elevation tiers (Control 4 / Inner 6 / Outer 8;
pills fully round), the spacing ramp, the type ramp (roles +
regular/semiBoldonly, no 500), two icon sizes (16 base / 12 compact), a 1px token for standard strokes,--vscode-*theme color, reveal-on-intent chrome, prefer a word to a mystery glyph.
Reach for a move only after naming the feeling and the principle - never instead. Describe a bug by its role/tier/ramp ("this overlay is rounded at the control tier"), not its number ("border-radius should be 6").
Full reference (values, principles, moves, worked examples, phrasebook, and
feedback guidance): the design-philosophy skill.
Token mechanics: design-tokens.instructions.md.
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.
- 2d ago First seen · 37 lines · 577 tokens per session scan A 6a326096f71e
vscode design-philosophy.instructions.md is an instructions file published in the GitHub repository microsoft/vscode (190,384 stars, last pushed today), licensed MIT. It adds 577 tokens to every session, about $0.0029 per session on Opus 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 instructions, from other repositories
open-codesign CLAUDE.md
Instructions for OpenCoworkAI/open-codesign, covering claude.md — open codesign, what this project is, hard constraints (do not violate), stack & conventions and frontend stack (locked).
mediago AGENTS.md
Instructions for mediago-dev/mediago, covering repository guidelines, project structure & module organization, tooling and script placement, documentation site safety and build, test, and development commands.
azure-sdk-for-go go-code.instructions.md
Instructions for Azure/azure-sdk-for-go: All code should follow the guidelines from the Azure Go SDK Guidelines. This document is a summary of the most important guidelines to follow when contributing to the Azure Go SDK.
drawio-ai-kit AGENTS.md
Instructions for sparklabx/drawio-ai-kit, covering repository guidelines, tech stack & runtime and code conventions & common patterns.
ChunUI CLAUDE.md
Instructions for liseami/ChunUI: Swift Package（swift-tools 6.2 · Swift 5 语言模式 + 默认 MainActor 隔离 + 渐进并发）+ iOS 18.6+ + 唯一依赖 Pow(MIT) 定位：从 Zinner（阿奇）提取的可换肤设计系统——宿主只定义 CCColors 调色板即得同质感全家桶；所有 shader/点阵动效随包。.
holaOS CLAUDE.md
Claude Code instructions for holaboss-ai/holaOS, covering holaos — repo guide and hola — debug the hola (pi) brain from the cli.