End-to-end tests for the Vibe CLI clients. Drives the built Rust CLI binary as a real user would: it sends terminal input and app-server JSON-RPC events, then compares the resulting terminal captures against committed goldens. - e2e/pty/ is product-agnostic. It owns PTY driving, key splitting, terminal emulation, and their data types; it must not import e2e.appserver. - e2e/appserver/ adapts scenarios, Vibe clients, and the replay app server to the generic terminal harness. App-server event builders belong here. - e2e/pty/capture.py takes…
A Rust/ratatui TUI for Vibe. It drives the unchanged Python engine over the app-server JSON-RPC protocol. Rewrite the vibe cli in rust for very good performances Committed Rust-only goldens (terminal SVGs + app-server request JSON) under client-e2e/goldens/<scenario>/. Generated by the existing e2e harness; no Python CLI needed to run the gate. Do not waste resources: when changes are focused, only run related test suites.
These instructions apply to AI-assisted work in this repository. - Start with README.md for project overview and install. - Use modelopt/ for source, tests/ for focused test coverage, and examples/ or docs/ for usage patterns. - Agent skills live under plugins/modelopt/skills/, the installable plugin's canonical skill tree. .agents/skills and .claude/skills expose those skills through relative symlinks. Shared agent config and scripts remain under .agents/. See .agents/README.md for the convention. - Coding guide: Code development and review require reading and following…
This repository contains a TypeScript/JavaScript project together with a lite Python SDK. These instructions are for agents like Codex to contribute changes. After your changes you must verify that everything still builds and tests pass. Execute the following commands from the repository root: If you have updated the VS Code extension, please run the extension tests with: Update the component specifications if you have updated the type annotations and documentations of the components. Use clear titles and summaries. Include relevant…
This directory contains a native implementation of the AI-DLC (AI-Driven Development Life Cycle) methodology that ships to many CLI harnesses — today Claude Code, Kiro CLI, Kiro IDE, Codex CLI, Cursor, opencode, and GitHub Copilot, and any capable CLI you port it to — from a single hand-authored source. The hand-authored source lives in core/ (harness-neutral) + harness/<name>/ (per-CLI surfaces); bun scripts/package.ts regenerates the dist/<harness>/ trees. The core uses the same building blocks in every harness:
This file provides context and instructions for AI coding agents working on the Microsoft Fluent UI Blazor components library. This is a .NET Blazor component library that provides Fluent UI design system components for Blazor applications. The library wraps Microsoft's official Fluent UI Web Components and provides additional components leveraging the Fluent Design System. Unit tests are in tests/Core/Components/ mirroring the component structure in src/Core/Components/. Follow Conventional Commits:
Instructions for AI agents working on Docker documentation. This site builds https://docs.docker.com/ using Hugo. The /manuals prefix is stripped from published URLs: content/manuals/desktop/install.md becomes /desktop/install/ on the live site. When writing internal cross-references in source files, keep the /manuals/ prefix in the path — Hugo requires the full source path. The stripping only affects the published URL, not the internal link target. Anchor links must exactly match the generated heading ID (Hugo lowercases and slugifies headings). Content in _vendor/ and…
Koog is a Kotlin multiplatform framework for building AI agents with graph-based workflows. It supports JVM and JS targets and integrates with multiple LLM providers (OpenAI, Anthropic, Google, OpenRouter, Ollama) and Model Context Protocol (MCP). The project follows a modular architecture with a clear separation of concerns: AIAgent — Main orchestrator that executes strategies in coroutine scopes, manages tools via ToolRegistry, runs features through AIAgentPipeline, and handles LLM communication via PromptExecutor. AIAgentStrategy — Graph-based execution logic that defines workflows as…
This file helps AI coding agents work effectively in the Cloudflare Workers SDK monorepo. Prefer authoritative configuration and documentation over copying details into this file: copied versions, rule lists, and counts become stale. - Use pnpm, not npm or yarn. - Use the Node.js and pnpm versions declared in package.json. - Install dependencies with pnpm install. - Run commands from the workspace root unless package documentation says otherwise. - Before changing a package, read its AGENTS.md if it has one.…
This package contains the private Flue application for agent-powered Workers SDK automation. Channels receive events from external services and dispatch specialized agents with narrowly scoped tools and runtime access. GitHub issue reproduction is the first implemented workflow, not the boundary of the application. Keep shared architecture generic enough for additional maintenance and triage workflows without prematurely abstracting code used by only one workflow. Run the focused checks from the repository root: Run pnpm prettify before committing. There is currently no…
Vite plugin for Cloudflare Workers development. Exports cloudflare() plugin factory from src/index.ts. ESM-only output. bin/cf-vite is an experimental, internal delegate binary spawned by Cloudflare's "cf-dev" parent process — NOT part of the plugin's public API and not meant for direct end-user invocation. It is the sibling of wrangler's cf-wrangler binary, and the two MUST keep a shared spawn contract so the parent can drive either impl interchangeably. - Verb dispatch. cf-vite <verb> [flags]. dev and build are the verbs today;…
Custom Vitest pool for running Worker tests inside the actual workerd runtime. Do NOT repeat root AGENTS.md content (pnpm, changesets, code style, etc.). - Runs in Node.js, part of the pool bundle (no separate package export) - Validates pool options (WorkersPoolOptionsSchema) and resolves the project's Worker configuration into Miniflare options - src/pool/plugin.ts provides the cloudflareTest() Vite plugin, which injects cloudflare:test resolution and sets resolve conditions (workerd, worker, browser) parseCustomPoolOptions() in src/pool/config.ts resolves at most one configuration file into a normalised…
OAuth-2.0-with-PKCE flow against Cloudflare's dash.cloudflare.com (or staging / custom-overridden) endpoints. Used by wrangler and (in future) other Cloudflare CLIs. Internal-only — published as prerelease: true. Pluggable credential persistence layer that
Wrangler-specific context only. See root AGENTS.md for monorepo conventions. Main CLI for Cloudflare Workers. ~2k-line yargs command tree in src/index.ts. Entry point is src/cli.ts (NOT src/index.ts). - No console.* — use logger singleton - No _dirname / _filename — use getBasePath() - No global fetch — use undici's fetch - Route Cloudflare v4 API calls through the existing cfetch helpers such as fetchResult, fetchListResult, and fetchPagedListResult when possible; do not use global or undici fetch directly. Prefer these helpers over…
Guidance for AI coding agents working in this repository. Human contributors should also read CONTRIBUTING.md and the documentation site which are the source of truth. The NVIDIA Container Toolkit is a collection of utilities that allow you to run GPU-accelerated containers. It is designed to work with any container runtime, e.g. docker, containerd, cri-o, podman. It is written primarily in Go. Historically, NVIDIA GPUs were injected into containers via an OCI preStart hook. The preStart hook would invoke the nvidia-container-cli…
Entry point for AI/LLM agents working in or with this repository. Pointers below; nothing canonical lives in this file. Reusable agent skills for working in this Nx workspace (Nx-related: nx-workspace, nx-generate, nx-import, nx-plugins, nx-run-tasks, link-workspace-packages) live at: Both directories currently hold the same content. Treat .agents/skills/ as the source of truth; the .claude/skills/ copy is kept in sync until Claude Code can discover skills from .agents/skills/ directly. Each package has its own README with usage examples and API notes: The…
MistralAGENTS.md
