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/dirac-run/dirac/agents-mdgit clone --depth 1 https://github.com/dirac-run/diracWhat 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 | $0.00777 | $0.00777 |
| Opus 5 | $0.00388 | $0.00388 |
| Sonnet 5 | $0.00155 | $0.00155 |
| Haiku 4.5 | $0.00078 | $0.00078 |
Grade A, and why
dirac AGENTS.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 — 51 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Dirac Agent Guide
This is the codebase of our coding agent Dirac. It supports cli and vscode extension.
Codebase Structure
src/core/task/: Task execution loop and statesrc/core/task/tools/modules/: Tool implementationssrc/core/task/tools/interfaces/IToolEnvironment.ts: Tool↔environment contractsrc/core/task/tools/ToolExecutorCoordinator.ts: Tool dispatchsrc/core/prompts/: System prompt templatessrc/core/controller/: Extension coordination and statesrc/core/context/: Context gatheringsrc/core/api/providers/: Per-provider API handlers (anthropic, gemini, openai, bedrock…)src/core/api/transform/: Provider stream → internalApiStreamsrc/shared/api.ts: Model IDs, pricing, capability flagssrc/integrations/: Terminal, Editor, Browser, and external service integrationssrc/services/: Shared services (logging, telemetry, tree-sitter, ripgrep, search, etc.)src/shared/: Cross-component types and utilitieswebview-ui/: React frontend (Cardis the UI primitive for tools)cli/: TypeScript/Ink CLIproto/dirac/: Protobuf definitions
Tooling
Tools are self-contained units dispatched by ToolExecutorCoordinator. Once called, a tool owns its lifecycle: it creates one or more Cards via createCard, updates their status via the card handle, and sets a final status before exiting. The only way a tool interacts with the environment is through IToolEnvironment. How cards are rendered is the UI's concern, not the tool's.
Must Follow Engineering Principles
- Layered architecture is a hard boundary. Task loop orchestrates, tools execute, UI renders, API handlers handler API providers. No layer knows the internals of another. Tool-specific logic belongs in the tool module, not in the coordinator or task loop. Never EVER mix the layers.
- No defensive programming, it does more harm than good to swallow failures.
- Every function should have single responsibility.
- No spaghetti control flow. Linear over nested. Early returns over deep conditionals. State transitions happen in one place.
- Tools are hermetically sealed. A tool's only interface with the environment is IToolEnvironment. Tools do not import each other. Shared behavior is explicit, not implicit coupling.
- Names are contracts. A function or file name should make its behavior predictable before you open it. Vague names (handle, process, manage) are rejected at review.
- No cargo-cult patterns. Don't add abstraction, indirection, or boilerplate because it feels right. Every layer of indirection needs a reason. Premature generalization is complexity debt.
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 · 51 lines · 777 tokens per session scan A f598c10a389e
dirac AGENTS.md is an instructions file published in the GitHub repository dirac-run/dirac (1,470 stars, last pushed 5d ago), licensed Apache-2.0. It adds 777 tokens to every session, about $0.0039 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
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.
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).
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.
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).
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.
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.