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/dubsopenhub/swarm-command/copilot-instructionsgit clone --depth 1 https://github.com/DUBSOpenHub/swarm-commandWhat 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 | $0.00650 | $0.00650 |
| Opus 5 | $0.00325 | $0.00325 |
| Sonnet 5 | $0.00130 | $0.00130 |
| Haiku 4.5 | $0.00065 | $0.00065 |
Grade A, and why
swarm-command 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.
How it starts
The opening of the file, as written. The whole thing — 44 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Copilot Instructions — Swarm Command
This repository contains Swarm Command, a GitHub Copilot CLI skill that orchestrates multi-model agent swarms with consensus scoring and shadow validation.
File map
| File/Dir | Purpose |
|---|---|
skills/swarm-command/SKILL.md |
Nexus orchestrator prompt (the brain) |
agents/swarm-command.agent.md |
Standalone agent version |
templates/*.md |
Prompt templates (Commander, Worker, Reviewer, Squad Lead) |
protocols/*.md |
Protocol invariants (Depth Guard, Circuit Breaker, Context Capsule) |
config.yml |
Tunables (models, thresholds, timeouts, scaling) |
catalog.yml |
Skill metadata + file references |
docs/*.md |
Architecture, consensus, shadow scoring, scaling docs |
Non-negotiables
- Depth Guard is sacred. Workers are ALWAYS leaf nodes (
explore/task). They MUST NOT use thetasktool. Three enforcement layers: prompt-level, contract-level, config-level. - Config is the source of truth. Never hardcode model names, thresholds, or tunables inside prompts.
- All outputs are strict JSON. Every layer boundary uses schema-validated JSON. No prose parsing.
- Just a skill. Do not add runtime code, package managers, telemetry, dashboards, or plugin systems.
- Parent controls spawning. The parent computes
can_launchfor every child. The child never decides for itself. - Sealed criteria are hidden. Shadow scoring uses the Shadow Score Spec sealed-envelope protocol. Sealed acceptance criteria are generated before commanders execute and must never appear in any agent prompt. Only the Nexus sees them during Phase 6 validation.
- Context compresses monotonically. 128K → 2K → 512 → 128 tokens. File scope narrows at every layer.
Prohibited actions
- Spawning
general-purposeagents at the worker level (depth 2+ for SS-50/SS-100, depth 3+ for SS-250) - Setting
can_launch = truefor Workers and Reviewers is prohibited. Squad Leads at depth 2 (SS-250 only) MAY havecan_launch=true(they spawn leaf workers only). At SS-50/SS-100, commanders spawn workers directly at depth 2. - Passing full context (128K) to workers — compress to 128 tokens
- Exposing shadow scoring criteria to workers or reviewers
- Hardcoding model names in prompt templates (use placeholders)
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 · 44 lines · 650 tokens per session scan A 0631f00a8f9c
swarm-command copilot-instructions.md is an instructions file published in the GitHub repository DUBSOpenHub/swarm-command (5 stars, last pushed 26d ago), licensed MIT. It adds 650 tokens to every session, about $0.0032 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 2026-08-31.
Other instructions, from other repositories
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
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.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
next.js AGENTS.md
Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.