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/ron537/dplex/copilot-instructionsgit clone --depth 1 https://github.com/Ron537/DPlexWhat 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 | $0.02614 | $0.02614 |
| Opus 5 | $0.01307 | $0.01307 |
| Sonnet 5 | $0.00523 | $0.00523 |
| Haiku 4.5 | $0.00261 | $0.00261 |
Grade A, and why
DPlex 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 yesterday.
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 — 165 lines — stays where its author put it; the contents beside it link to each section on GitHub.
DPlex — Copilot Instructions
Build & Run
npm run dev # Start Electron app in dev mode (hot reload)
npm run build # Typecheck + build (runs typecheck:node && typecheck:web first)
npm run lint # ESLint across entire project
npm run typecheck # TypeScript check only (no build)
npm run test:unit # Run unit tests (Vitest)
npm run test:e2e # Run Electron end-to-end tests (Playwright)
npm run test:monkey # Run randomized monkey tests (Playwright)
npm run test:all # Run all tests
Architecture
DPlex is an Electron + React terminal multiplexer that manages AI CLI tool sessions (Copilot CLI, Claude Code, etc.). It uses electron-vite with three process targets:
Process Boundaries
- Main process (
src/main/) — Node.js. PTY management vianode-pty, session discovery from filesystem, settings persistence. All file system and process operations live here. - Preload (
src/preload/index.ts) — IPC bridge. DefinesDplexAPIinterface exposed aswindow.dplex. This is the sole contract between main and renderer — every new IPC channel must be added to both the type definition and the implementation in this file. - Renderer (
src/renderer/) — React + Tailwind CSS v4. No direct Node.js or filesystem access; everything goes throughwindow.dplex.*IPC calls.
Provider System (AI Tool Abstraction)
AI tools are abstracted behind SessionProvider interface (src/main/services/providers/types.ts). To add a new AI CLI tool:
- Create a class implementing
SessionProviderinsrc/main/services/providers/ - Register it in
createDefaultRegistry()inproviders/index.ts
No other files need changes. The provider handles: session discovery, active session detection, close/delete, resume command generation, and session resolution by PID/CWD.
ProviderRegistry aggregates operations across all providers. IPC handlers in main/index.ts delegate to the registry.
State Management
Zustand stores (no Redux). Four stores, each owning a specific domain:
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.
- yesterday First seen · 165 lines · 2,614 tokens per session scan A 2c5608f76db9
DPlex copilot-instructions.md is an instructions file published in the GitHub repository Ron537/DPlex (18 stars, last pushed 8d ago), licensed MIT. It adds 2,614 tokens to every session, about $0.0131 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.
Other instructions, from other repositories
skill-recorder copilot-instructions.md
Instructions for microsoft/skill-recorder, covering repository instructions for coding agents and pull requests target the active release branch, not main.
gortex CLAUDE.md
Instructions for zzet/gortex, covering gortex, build & test, codebase overview, discovery (read once, then keep using) and llm provider (powers ask and searchsymbols assist: modes).
calva-backseat-driver AGENTS.md
Instructions for BetterThanTomorrow/calva-backseat-driver, covering calva backseat driver - ai agent guide, essential architecture, action/effect system ("ex"), runtime state structure and datascript connections.
calva-backseat-driver backseat-driver-source-conventions.instructions.md
Backseat Driver extension source code conventions — Ex framework patterns, async, state management, namespace separation, and naming. Supplements universal Clojure patterns with Backseat Driver-specific rules.
copilotd copilot-instructions.md
Instructions for DamianEdwards/copilotd, covering copilot instructions for copilotd, build & run, build, run from source (auto-builds and passes all args through) and architecture.
calva-backseat-driver copilot-instructions.md
Instructions for BetterThanTomorrow/calva-backseat-driver, covering what backseat driver is, repl environment, state inspection safety, bundled ai context (chatskills / chatinstructions) and key development commands.