vscode-copilot-chat model-prompts.instructions.md

Guidelines for creating prompts tailored to different AI models in Copilot Chat. A prompt registry chooses the first matching model handler or falls back to default settings.

In plain words
What is it for?
Adding, registering, or testing model-specific prompts and their system or reminder instructions in Copilot Chat.
Why use it?
They explain how model-specific prompt handlers are selected, so new handlers do not get hidden by broader matches or incorrect registration order.

Instructions file for GitHub Copilot

Install

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.

agentmods
npx agentmods add instructions/microsoft/vscode-copilot-chat/model-prompts
Clone the repo
git clone --depth 1 https://github.com/microsoft/vscode-copilot-chat

Made for: GitHub Copilot.

Per session 1,649 This file is loaded in full into every session.
When invoked 1,649 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.01649 $0.01649
Opus 5 $0.00825 $0.00825
Sonnet 5 $0.00330 $0.00330
Haiku 4.5 $0.00165 $0.00165

Measured 2d ago against content hash ddc8204d7181, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

vscode-copilot-chat 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 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.

.github/instructions/model-prompts.instructions.md · 194 lines

How it starts

The opening of the file, as written. The whole thing — 194 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:

  1. Phase 1 — matchesModel(): Iterates registered resolvers in order, calling matchesModel(). First true wins.
  2. Phase 2 — familyPrefixes: If no matchesModel matched, checks endpoint.family.startsWith(prefix). First match wins.
  3. Defaults: If no resolver matches, defaults are used for all customizations.

Registration order matters — more specific resolvers (e.g., hash-based) should be registered before broader ones (e.g., prefix-based fallbacks).

Resolver Interface

The IAgentPrompt interface provides these customization methods:

Method Returns Purpose
resolveSystemPrompt(endpoint) SystemPrompt Main system prompt class
resolveReminderInstructions?(endpoint) ReminderInstructionsConstructor Reminder instructions appended to user messages
resolveToolReferencesHint?(endpoint) ToolReferencesHintConstructor Tool usage hints
resolveCopilotIdentityRules?(endpoint) CopilotIdentityRulesConstructor Identity/role rules
resolveSafetyRules?(endpoint) SafetyRulesConstructor Safety rules
resolveUserQueryTagName?(endpoint) string Tag name wrapping user queries

All methods are optional except resolveSystemPrompt. Unimplemented methods fall back to defaults (DefaultReminderInstructions, DefaultToolReferencesHint, etc.).

Dependency Injection in Resolvers

Resolvers are created via instantiationService.createInstance(), so they support full constructor DI. Inject IConfigurationService, IExperimentationService, etc. to read experiment flags or config values:

Read the full file on GitHub · 194 lines

Changes

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.

  1. 2d ago First seen · 194 lines · 1,649 tokens per session scan A ddc8204d7181

Subscribe to this mod's changes

vscode-copilot-chat model-prompts.instructions.md is an instructions file published in the GitHub repository microsoft/vscode-copilot-chat (9,973 stars, last pushed 3mo ago), licensed MIT. It adds 1,649 tokens to every session, about $0.0082 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.