agentmods.dev

drift copilot-instructions.md

drift copilot-instructions.md is an instructions file for GitHub Copilot from mick-gsk/drift. It costs 3,275 tokens per session, scanned A, original, MIT.

A required set of working rules for AI coding assistants in the Drift workspace. It tells agents how to handle prompts, instructions, skills, and agent files.

In plain words
What is it for?
Following Drift's policies while editing prompt and agent configuration files. It helps choose the right kind of instruction file and apply required prompt-writing rules.
Why use it?
It gives every agent the same rules and prevents inconsistent or weak instructions. It also directs agents to read the workspace policy before starting work.

Instructions file for GitHub Copilot

not rated 14 1mo ago A scan 3,275 tokens original MIT
View the source file mick-gsk/drift
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/mick-gsk/drift/copilot-instructions
Clone the repo
git clone --depth 1 https://github.com/mick-gsk/drift

Made for: GitHub Copilot.

Wrote 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.

agentmods badge for drift copilot-instructions.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/mick-gsk/drift/copilot-instructions.svg)](https://agentmods.dev/instructions/mick-gsk/drift/copilot-instructions)
Your own site
<a href="https://agentmods.dev/instructions/mick-gsk/drift/copilot-instructions"><img src="https://agentmods.dev/badge/instructions/mick-gsk/drift/copilot-instructions.svg" alt="Measured on agentmods" height="20"></a>
Per session 3,275 This file is loaded in full into every session.
When invoked 3,275 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 $0.03275 $0.03275
Opus 5 $0.01638 $0.01638
Sonnet 5 $0.00655 $0.00655
Haiku 4.5 $0.00328 $0.00328

Measured 3d ago against content hash 269e00e2f90a, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

drift 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 3d 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 · 277 lines

How it starts

The opening of the file, as written. The whole thing — 277 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Drift — Verbindliche Arbeitsgrundlage für alle Agenten

Diese Datei ist für alle Copilot-Agenten, Coding-Agenten und KI-Assistenten im Drift-Workspace bindend.

Die vollständige Policy befindet sich in: POLICY.md (Workspace-Root)

Lies diese Datei vor jeder Arbeit vollständig, sofern sie nicht bereits im Kontext ist. Die Policy ist ein Vertrag — keine Empfehlung, kein Vorschlag.

Primärmodus bei Prompt-Engineering

Wenn eine Aufgabe Prompts, Instructions, Skills, Agents oder diese Datei selbst betrifft, arbeitet der Agent im Prompt-Engineering-Modus. Ziel ist nicht schönere Prosa, sondern härtere, operativere und reviewbare Agentensteuerung.

Betroffene Pfade:

  • .github/prompts/**
  • .github/instructions/**
  • .github/skills/**
  • .github/agents/**
  • .github/AGENTS.md
  • .github/copilot-instructions.md

Detaillierte Zusatzregeln:

  • Routing: .github/instructions/drift-context-routing.instructions.md
  • Prompt-Engineering: .github/instructions/drift-prompt-engineering.instructions.md
  • Workflow-Skill: .github/skills/drift-agent-prompt-authoring/SKILL.md

Zuerst das richtige Primitive wählen

Bedarf Richtiges Primitive
Repo-weite, immer geltende Regeln copilot-instructions.md
Datei- oder Ordner-spezifische Regeln *.instructions.md
Wiederverwendbarer Operator-Workflow SKILL.md
Konkreter mehrphasiger Ablauf mit Ziel und Artefakten *.prompt.md
Isolierter Spezialmodus mit eigener Tool-/Kontextgrenze *.agent.md

Falsches Primitive = schwacher Prompt. Eine Datei darf nur das regeln, wofür ihr Primitive gedacht ist.

Nicht verhandelbare Schärferegeln

  1. Deutsch und modellunabhängig. Keine Modellnamen, keine vendor-spezifischen Prompt-Tricks.
  2. Description ist Discovery-Oberfläche. Das description-Feld muss Triggerwörter, Task-Typ und Scope explizit enthalten.
  3. applyTo so eng wie möglich. applyTo: "**" nur für wirklich universelle Regeln.
  4. Ein Problem pro Datei. Keine Mischdateien für Policy, Testing, Release und Prompt-Stil.
  5. Keine Parallel-Policy. Shared Partials, Policy und Push-Gates referenzieren statt duplizieren.
  6. Keine weichen Verben ohne Vertrag. Wörter wie "analysiere", "verbessere", "prüfe gründlich", "wenn möglich" oder "robust" sind nur zulässig, wenn Inputs, Schritte, Ergebnisform und Abbruchkriterium definiert sind.
  7. Keine Halluzinationsflächen. Keine erfundenen Flags, Dateien, Tools, Issue-Ziele, Pfade oder angeblich vorhandenen Artefakte.
  8. Prompts erzeugen Evidenz. Jeder Prompt muss in Beobachtung, Artefakt, Entscheidung, Maßnahme oder sauberem Abbruch enden - nie nur in Prosa.

Read the full file on GitHub · 277 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. 3d ago First seen · 277 lines · 3,275 tokens per session scan A 269e00e2f90a

Subscribe to this mod's changes

drift copilot-instructions.md is an instructions file published in the GitHub repository mick-gsk/drift (14 stars, last pushed 1mo ago), licensed MIT. It adds 3,275 tokens to every session, about $0.0164 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.

ai-generated-codearchitectureclicode-qualitycopilot
Related

Other instructions, from other repositories

truecourse CLAUDE.md

Claude Code instructions for truecourse-ai/truecourse, covering truecourse — claude instructions, key files to keep updated, project layout, development commands and storage.

truecourse-ai/truecourse · 7,282 tokens

PhpCodeArcheology CLAUDE.md

Instructions for PhpCodeArcheology/PhpCodeArcheology, covering claude.md — phpcodearcheology, project overview, language, tech stack and project structure.

PhpCodeArcheology/PhpCodeArcheology · 976 tokens

vibe-check AGENTS.md

Instructions for benavlabs/vibe-check, covering security rules, secrets, database, authentication and authorization and input and output.

benavlabs/vibe-check · 1,071 tokens

plan-forge copilot-instructions.md

Copilot instructions for srnichols/plan-forge, covering instructions for copilot, architecture principles, core principles (non-negotiable), cost estimates and talking to forge-master.

srnichols/plan-forge · 2,670 tokens

plan-forge auth.instructions.md

.NET authentication & authorization — JWT/OIDC, policy-based auth, multi-tenant isolation, API keys, testing.

srnichols/plan-forge · 2,552 tokens

plan-forge blazor-fluent-ui.instructions.md

Blazor Server + Microsoft Fluent UI patterns — component layering, state, lifecycle, accessibility. Auto-loads when editing .razor / .razor.cs files.

srnichols/plan-forge · 3,140 tokens