Visual Studio Code is a code editor that supports editing, navigating, understanding, debugging, and extending software projects. Developers use it for the edit-build-debug cycle, and the catalogue add-ons provide skills, instructions, and agents for working within the editor.
Borrowing it
Nothing to install: this file belongs to microsoft/vscode. 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/microsoft/vscode/main/extensions/copilot/.github/instructions/model-prompts.instructions.mdgit clone --depth 1 https://github.com/microsoft/vscodeWrote 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/microsoft/vscode/model-prompts)<a href="https://agentmods.dev/instructions/microsoft/vscode/model-prompts"><img src="https://agentmods.dev/badge/instructions/microsoft/vscode/model-prompts/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/microsoft/vscode/model-prompts"><img src="https://agentmods.dev/badge/instructions/microsoft/vscode/model-prompts.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 | $0.02022 | $0.02022 |
| Opus 5.5 | $0.00809 | $0.00809 |
| Sonnet 5.5 | $0.00404 | $0.00404 |
| Haiku 4.5 | $0.00202 | $0.00202 |
Grade A, and why
vscode model-prompts.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 1mo 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 — 206 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Guidelines for creating, registering, and testing model-specific prompts via the PromptRegistry system.
Prompt Registry
The PromptRegistry in promptRegistry.ts maps AI models to their optimal prompt structures. Each model provider has a resolver class implementing IAgentPrompt that returns prompt customizations for that provider's models.
Resolution Order
When PromptRegistry.resolveAllCustomizations() is called:
- Phase 1 —
matchesModel(): Iterates registered resolvers in order, callingmatchesModel(). Firsttruewins. - Phase 2 —
familyPrefixes: If nomatchesModelmatched, checksendpoint.family.startsWith(prefix). First match wins. - Phase 3 — provider fallback: If neither phase matched, checks predicates registered through
registerFallbackPrompt(resolver, matchesModel). First match wins. - Defaults: If no resolver matches, defaults are used for all customizations.
Registration order matters within each phase. Register broad provider defaults through registerFallbackPrompt, not matchesModel or familyPrefixes, so they never shadow a model-specific resolver or a family-prefix registration.
OpenAI Default
Unrecognized GPT families and endpoints identified as OpenAI use the latest approved GPT prompt bundle. The single promotion point is the LatestOpenAIPromptResolver import in openai/latestOpenAIPrompt.ts, currently GPT-6 Astra. It reuses the entire resolver, including reminders, identity, safety, tool-reference hints, and user-query tag defaults, without copying prompt text.
- Do not add a prompt file just because a new OpenAI model ships. Add a dedicated resolver only when the model needs different instructions.
- Existing specialized GPT/Codex resolvers and the explicitly retained legacy families in
defaultOpenAIPrompt.tsxtake precedence. Version matchers must distinguishgpt-5.1and its hyphenated variants fromgpt-5.10. - OpenAI provider metadata can identify opaque preview names; an OpenAI-compatible API alone does not establish OpenAI model identity. Explicit family routing takes precedence over the provider fallback.
- Prompt inheritance must not alias the endpoint family or enable GPT-6-specific API/tool capabilities. Rendering still respects the actual endpoint's capabilities and available tools.
- Hidden model I is explicitly matched by the GPT-6 resolver and included alongside GPT-6 capability and instruction gates, independently of the provider fallback. Its model ID and family remain unchanged.
- To promote a new default, update the resolver import in
latestOpenAIPrompt.tsand the expected default intest/openAIPrompts.spec.ts. Run its routing/rendering tests and the existing agent prompt snapshots. Keep older specialized resolvers unchanged.
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.
- 1mo ago Changed · +1 lines · +35 tokens per session b16ca940e882
- 1mo ago First seen · 205 lines · 1,987 tokens per session scan A 5fad036783cc
vscode model-prompts.instructions.md is an instructions file published in the GitHub repository microsoft/vscode (194,219 stars, last pushed today), licensed MIT. It adds 2,022 tokens to every session, about $0.0081 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 2026-09-03.
Other instructions, from other repositories
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
deno copilot-instructions.md
Copilot instructions for denoland/deno, covering deno development guide for github copilot, network access, high level overview, key directories and key files to understand first.
deno CLAUDE.md
Claude Code instructions for denoland/deno, covering deno development guide, git workflow, high level overview, key directories and quick start.
generative-ai-for-beginners AGENTS.md
AGENTS.md instructions for microsoft/generative-ai-for-beginners, covering agents.md, project overview, setup commands, initial repository setup and clone the repository.