Show HN: VT Code – open-source terminal coding agent in Rust
Recorded: May 30, 2026, 5:01 a.m.
| Original | Summarized |
GitHub - vinhnx/VTCode: VT Code is an open-source coding agent with LLM-native code understanding and robust shell safety. Supports multiple LLM providers with automatic failover and efficient context management. · GitHub Skip to content Navigation Menu Toggle navigation
Sign in
Appearance settings PlatformAI CODE CREATIONGitHub CopilotWrite better code with AIGitHub SparkBuild and deploy intelligent appsGitHub ModelsManage and compare promptsMCP RegistryNewIntegrate external toolsDEVELOPER WORKFLOWSActionsAutomate any workflowCodespacesInstant dev environmentsIssuesPlan and track workCode ReviewManage code changesAPPLICATION SECURITYGitHub Advanced SecurityFind and fix vulnerabilitiesCode securitySecure your code as you buildSecret protectionStop leaks before they startEXPLOREWhy GitHubDocumentationBlogChangelogMarketplaceView all featuresSolutionsBY COMPANY SIZEEnterprisesSmall and medium teamsStartupsNonprofitsBY USE CASEApp ModernizationDevSecOpsDevOpsCI/CDView all use casesBY INDUSTRYHealthcareFinancial servicesManufacturingGovernmentView all industriesView all solutionsResourcesEXPLORE BY TOPICAISoftware DevelopmentDevOpsSecurityView all topicsEXPLORE BY TYPECustomer storiesEvents & webinarsEbooks & reportsBusiness insightsGitHub SkillsSUPPORT & SERVICESDocumentationCustomer supportCommunity forumTrust centerPartnersView all resourcesOpen SourceCOMMUNITYGitHub SponsorsFund open source developersPROGRAMSSecurity LabMaintainer CommunityAcceleratorGitHub StarsArchive ProgramREPOSITORIESTopicsTrendingCollectionsEnterpriseENTERPRISE SOLUTIONSEnterprise platformAI-powered developer platformAVAILABLE ADD-ONSGitHub Advanced SecurityEnterprise-grade security featuresCopilot for BusinessEnterprise-grade AI featuresPremium SupportEnterprise-grade 24/7 supportPricing Search or jump to... Search code, repositories, users, issues, pull requests...
Search Clear
Search syntax tips Provide feedback Include my email address so I can be contacted Cancel Submit feedback Saved searches
Name Query To see all available qualifiers, see our documentation. Cancel Create saved search Sign in Sign up
Appearance settings Resetting focus You signed in with another tab or window. Reload to refresh your session. Dismiss alert vinhnx VTCode Public Uh oh! There was an error while loading. Please reload this page.
Notifications
Fork
Star Code Issues Pull requests Discussions Actions Projects Wiki Security and quality Insights
Additional navigation options
Code Issues Pull requests Discussions Actions Projects Wiki Security and quality Insights
mainBranchesTagsGo to fileCodeOpen more actions menuFolders and filesNameNameLast commit messageLast commit dateLatest commit History5,767 Commits5,767 Commits.github.github .vtcode.vtcode docsdocs evalsevals fuzzfuzz homebrewhomebrew resourcesresources rulesrules scriptsscripts srcsrc system-promptssystem-prompts teststests vscode-extensionvscode-extension vtcode-acpvtcode-acp vtcode-authvtcode-auth vtcode-bash-runnervtcode-bash-runner vtcode-collaboration-tool-specsvtcode-collaboration-tool-specs vtcode-commonsvtcode-commons vtcode-configvtcode-config vtcode-corevtcode-core vtcode-exec-eventsvtcode-exec-events vtcode-file-searchvtcode-file-search vtcode-ghostty-vt-sysvtcode-ghostty-vt-sys vtcode-indexervtcode-indexer vtcode-llmvtcode-llm vtcode-lmstudiovtcode-lmstudio vtcode-markdown-storevtcode-markdown-store vtcode-process-hardeningvtcode-process-hardening vtcode-terminal-detectionvtcode-terminal-detection vtcode-themevtcode-theme vtcode-toolsvtcode-tools vtcode-tuivtcode-tui vtcode-utility-tool-specsvtcode-utility-tool-specs vtcode-vimvtcode-vim zed-extensionzed-extension .env.example.env.example .gitignore.gitignore .mcp.json.mcp.json .python-version.python-version .vtcodegitignore.vtcodegitignore AGENTS.mdAGENTS.md CHANGELOG.mdCHANGELOG.md CONTEXT.mdCONTEXT.md Cargo.lockCargo.lock Cargo.tomlCargo.toml Cross.tomlCross.toml Dockerfile.buildDockerfile.build LICENSELICENSE README.mdREADME.md build.rsbuild.rs cliff.tomlcliff.toml clippy.tomlclippy.toml config.tomlconfig.toml release.tomlrelease.toml rust-toolchain.tomlrust-toolchain.toml rustfmt.tomlrustfmt.toml sgconfig.ymlsgconfig.yml vtcode.tomlvtcode.toml vtcode.toml.examplevtcode.toml.example View all filesRepository files navigationREADMECode of conductContributingMIT licenseSecurity VT Code is an open-source coding agent with LLM-native code understanding and robust shell safety. Supports multiple LLM providers with automatic failover and efficient context management. Installation macOS & Linux: # Or skip ripgrep + ast-grep during install Alternative Installation Methods: # Homebrew (macOS/Linux) # From vinhnx/tap (development/bleeding edge): # If switching between taps, uninstall first: # Optional: install the search tools bundle later Bundled in official macOS/Linux release archives under ghostty-vt/ See Installation Guide, Development Setup, and Ghostty VT Packaging. Note: If you encounter "installed from different tap" errors, uninstall the existing formula first: brew uninstall vtcode, then install from your preferred tap. See Installation Guide and Native Installer Guide for more options and troubleshooting. # Launch VT Code Skill Discovery: VT Code can discover and load skills from multiple locations with precedence handling For more information about skills implementation in VT Code, see Agent Skills Guide. /agent and /agents active inspect delegated agents in place without switching the main session The background runtime is configured under [subagents.background]: [agent] [[custom_providers]] # Authenticate # Or use VT Code's TUI Agent Discovery: Via Agent Cards at /.well-known/agent-card.json For more information about A2A protocol support, see A2A Protocol Documentation. Messages endpoint at /v1/messages For more information about Anthropic API compatibility, see Anthropic API Documentation. Unified Item Types: State machine-based items with defined lifecycle states For more information about Open Responses implementation, see Open Responses Documentation. Complete History: Every agent step, tool invocation, and observation in one file For more information, see ATIF Trajectory Format Documentation. OAuth Providers: OpenAI ChatGPT and OpenRouter with PKCE-secured flows For detailed setup and provider guides, see OAuth Authentication Guide. OAuth Settings: Control authentication method and token storage per provider For full configuration options, see Configuration. Output Streams: Primary output (e.g., code from ask, events from exec) is sent to stdout. All logs, metadata, reasoning traces, and interaction prompts are sent to stderr. This allows you to safely pipe vtcode output to other tools. Standard Arguments: Adheres to GNU/POSIX argument parsing standards using clap, supporting standard flags like --help, --version, and -- separator for positional arguments. Key Features Security First: Multi-layered security model with tree-sitter-bash command validation, execution policy, and sandbox isolation Security & Safety Execution Policy: Command allowlist with per-command argument validation See Security Model and Sandbox Deep Dive for details. Docs & Examples Installation Native Installers Getting started Interactive mode OAuth Authentication - Secure OAuth 2.0 flows with PKCE, token storage, and automatic refresh OpenAI, Anthropic, Google Gemini Context Engineering Token budget management Agent Skills - Comprehensive skills location system with multi-location support, precedence handling, and the open Agent Skills standard. Standard Compliance - How VT Code implements the Agent Skills standard Code Intelligence Development Setup FAQ - Frequently asked questions about VT Code Event Handling - Terminal event architecture and patterns Ratatui Integration - Complete integration of Ratatui FAQ best practices Summary - Overview of all improvements Architecture Security Model Anthropic API Compatibility - Documentation for Anthropic Messages API compatibility server Documents Ask Google Gemini CodeWiki Visual Studio Code Extension The original VT Code extension for Visual Studio Code with full semantic code understanding and AI assistance. Compatible with Cursor, Windsurf, and other VS Code-compatible editors through the Open VSX registry. Contributing Report issues you're experiencing If you're not sure where to start: Check out the issues page Steps to get started: Fork the repository by clicking the Fork button in the top-right corner When reporting an issue, please include enough details for others to reproduce the problem effectively. Star History License About VT Code is an open-source coding agent with LLM-native code understanding and robust shell safety. Supports multiple LLM providers with automatic failover and efficient context management. vinhnx.github.io Topics agent rust cli terminal tui cargo crossterm ratatui codingagent Resources Readme MIT license Code of conduct Code of conduct Contributing Contributing Security policy Security policy Uh oh! There was an error while loading. Please reload this page. Activity 618 6 51 Report repository Releases 0.115.0 Latest Sponsor this project
Uh oh! There was an error while loading. Please reload this page. patreon.com/vinhnx opencollective.com/vinhnx buymeacoffee.com/vinhnx Learn more about GitHub Sponsors Uh oh! There was an error while loading. Please reload this page. Contributors Uh oh! There was an error while loading. Please reload this page. Languages Rust TypeScript Shell Python PowerShell JavaScript
Footer © 2026 GitHub, Inc. Footer navigation Terms Privacy Security Status Community Docs Contact Manage cookies Do not share my personal information You can’t perform that action at this time. |
VT Code is presented as an open-source coding agent distinguished by its LLM-native code understanding and a robust shell safety framework. The system is designed to support multiple large language model providers through automatic failover mechanisms and efficient context management. The installation process offers native installers for macOS and Linux, alongside alternative methods using Cargo and Homebrew, with specific considerations for runtime library packaging, such as using Ghostty VT, which provides richer PTY screen snapshots. The agent incorporates an extensible Skill system, adhering to the Agent Skills specification, which allows it to discover and load skills from various locations, handling precedence correctly for interoperability. It manages internal execution through the delegation of bounded work to foreground subagents and explicitly configured background subagents, accessible via commands like /agent and /subprocesses, which are unified in a wide-mode sidebar. The system supports a diverse range of AI providers, including OpenAI, Anthropic, Google Gemini, DeepSeek, and various inference providers like Ollama and LM Studio, facilitated by environment variable configuration. Custom integrations are supported via custom providers, exemplified by the Atlas Cloud integration. Interoperability is foundational to VT Code, as it implements several agent protocols. It supports the Agent Client Protocol for integration with code editors like Zed, and the Agent2Agent Protocol, which enables communication and state management between different AI agents through features like agent discovery, task lifecycle management, real-time streaming via Server-Sent Events, and JSON-RPC 2.0 for general interoperability. Furthermore, VT Code achieves compatibility with the Anthropic Messages API, supporting features such as tool calling, vision support, and multi-turn conversations, and conforms to the Open Responses specification to ensure vendor-neutral LLM workflow capabilities. Session history and debugging are formalized through the Agent Trajectory Interchange Format (ATIF) v1.4 implementation. ATIF standardizes the export of complete interaction histories, encompassing user messages, agent responses, tool calls, observations, and LLM metrics like token usage, thereby enabling standardized pipelines for debugging, synthetic data generation, and reinforcement learning. Security is a paramount concern for VT Code, employing a multi-layered defense-in-depth security model. This model includes execution policy with command allowlists and per-command argument validation, workspace isolation, and OS-native sandboxing leveraging security mechanisms such as macOS Seatbelt and Linux Landlock and seccomp for kernel-enforced isolation. The system also incorporates human-in-the-loop approval for sensitive operations and maintains a comprehensive audit trail of all command executions. The system follows strict command line interface design principles, adopting UNIX standards by routing primary output streams (like generated code) to stdout while channeling metadata, reasoning traces, and interaction prompts to stderr, promoting safe piping of output to other utilities. Configuration is managed via a configuration file, storing preferences in vtcode.toml, which governs OAuth settings, lifecycle hooks, tool policies, performance tuning parameters, and backend selection, defaulting to the Ghostty VT runtime library. The system features advanced context management through dynamic context curation and token budget tracking. It is integrated into development workflows via editor extensions, including a native Visual Studio Code extension and compatibility with other VS Code-compatible editors through the Open VSX registry. In terms of interaction, the system features a rich terminal user interface and supports agent events that can trigger custom shell commands via lifecycle hooks for contextual enrichment and automation. |