prompty copilot-instructions.md

Instructions for working with Prompty v2, a system where .prompty files combine configuration in a header with written instructions for running an AI prompt.

In plain words
What is it for?
Use them when changing Prompty's prompt format, runtime, model configuration, template and parser settings, inputs and outputs, or integrations with local functions, MCP servers, and OpenAPI services.
Why use it?
They explain which generated model types the runtime uses and how prompts, models, connections, templates, properties, and tools fit together.

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/prompty/copilot-instructions
Clone the repo
git clone --depth 1 https://github.com/microsoft/prompty

Made for: GitHub Copilot.

Per session 14,642 This file is loaded in full into every session.
When invoked 14,642 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 $0.14642 $0.14642
Opus 5 $0.07321 $0.07321
Sonnet 5 $0.02928 $0.02928
Haiku 4.5 $0.01464 $0.01464

Measured 2d ago against content hash 056659584066, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

prompty copilot-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/copilot-instructions.md · 1,450 lines

How it starts

The opening of the file, as written. The whole thing — 1,450 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Prompty v2 — Complete Rebuild Plan

Overview

Prompty is being rebuilt from the ground up. The type system that originated in Prompty's TypeSpec specification is now owned in this repo and emitted through Typra. Prompty v2 uses the generated Prompty model packages as the canonical, lower-level type layer.

Core principle: Prompty is a markdown file format (.prompty) for LLM prompts. The frontmatter uses Prompty's Typra-generated prompt model. The markdown body becomes instructions. The runtime loads, renders, parses, and executes prompts.

What Prompty Uses from the Typra-Generated Model

Only the prompt runtime subset — not every schema contract:

Model Type Prompty Uses
Prompty Yes — the loaded .prompty file becomes this
Model Yes — model configuration
Connection (all subtypes) Yes — auth/endpoint config
ModelOptions Yes — temperature, maxOutputTokens, etc.
Template / FormatConfig / ParserConfig Yes — template engine config
Property Yes — inputs/outputs
FunctionTool Yes — local function calling
McpTool Yes — MCP server integration
OpenApiTool Yes — API-based tools
CustomTool Yes — wildcard * catch-all for unknown tool kinds
Workflow No
ContainerAgent No
AgentManifest No
WebSearchTool / FileSearchTool / CodeInterpreterTool No

Read the full file on GitHub · 1,450 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 · 1,450 lines · 14,642 tokens per session scan A 056659584066

Subscribe to this mod's changes

prompty copilot-instructions.md is an instructions file published in the GitHub repository microsoft/prompty (1,254 stars, last pushed 4d ago), licensed MIT. It adds 14,642 tokens to every session, about $0.0732 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.