# April 2025 Source: https://docs.zencoder.ai/changelog/april-2025 Enhanced tooling capabilities, new Bash tool, Requirements tool, and significant UX improvements # Enhanced Developer Experience with New Tools and Improved Workflows **Important Update**: Zencoder now requires JetBrains IDE version 2024.2 or later to run. April brought significant improvements to Zencoder's functionality and user experience. We've focused on adding powerful new tools like the Bash/Shell tool and Requirements tool, enhancing UX, and making many under-the-hood improvements to make your development workflow more efficient. Here's what we've been working on: ## New Tools and Features ### Bash/Shell Tool for Coding Agent The Bash tool is now available to the coding agent with significant improvements: * Improved UX for safe commands that can run without confirmation * Management options including "Don't ask again" in both VS Code and JetBrains * Works seamlessly with Coffee Mode for increased productivity and speed * Enhanced MCP management of environment variables and npx, making it more stable Explore how the Coding Agent can now leverage shell commands to better assist your development workflow. ### Requirements Tool Our new Requirements tool helps the coding agent better understand your needs: * Automatically asks for clarifications when needed * Gives you greater control over your project and work * Better understands prompts and delivers higher-quality results * Reduces back-and-forth by getting requirements right the first time ### Enhanced Chat Experience We've made several improvements to the chat interface: * Added support and hints for using slash commands (`/`) and file mentions (`@`) in the chatbox * New warning system when chats become too long, helping you optimize token usage across multiple chats * Ability to attach files from outside of the project to provide better context ## Debugging and Support Improvements We've made it easier to get help when you need it: * Simplified access to Operation IDs and debug settings to help us troubleshoot issues more effectively **Helping Us Help You**: Operation IDs help our team trace exactly what happened during your interaction with the system. Learn how to access and share these IDs in our [Debug Information Guide](/user-guides/troubleshooting/debug-information#helping-us-help-you). ## Search and Review Enhancements * Significant improvements to local (client-side) search functionality * Enhanced `/review` command is now invoking a pre-built custom agent with the same name, providing more precise and valuable insights * You can adjust this custom agent's details in the Custom Agent settings Discover how to customize the review agent and other specialized agents for your workflow. ## Documentation Updates * New descriptions and updated GIFs for documentation on both JetBrains and VS Code marketplaces * Added detailed [guides on how to collect debug information and Operation IDs](/user-guides/troubleshooting/debug-information) to help with troubleshooting * Enhanced documentation to make it easier to get support when needed ## Bugs and Fixes We've addressed several user-facing issues: * Fixed an annoying bug where indexing sometimes gets stuck at "Updating index" after reopening the IDE * Fixed a problem with cursor positioning above inserted code snippets * Improvements to Jira authentication flow * Fixed hanging errors between the bash/shell tool and our internal processing *Note: We've also made many background improvements and fixes that aren't listed here but contribute to a more stable and reliable experience.* ## Version History * 1.28.0 (Apr 28, 2025) * 1.26.0 (Apr 17, 2025) * 1.24.0 (Apr 10, 2025) * 1.22.0 (Apr 2, 2025) * 1.24.0 (Apr 30, 2025) * 1.23.1 (Apr 24, 2025) * 1.23.0 (Apr 22, 2025) * 1.22.0 (Apr 15, 2025) * 1.21.0 (Apr 4, 2025) # Product Updates Source: https://docs.zencoder.ai/changelog/home Follow the ongoing evolution of our AI coding assistant # Intro We built this changelog to maintain transparency with you—our users and community. It documents our journey to create a more intelligent, helpful coding companion for your daily work. While we ship improvements weekly, we compile these updates into monthly changelog entries to provide an organized overview. This approach gives you a clear record of our progress while making it easier to track significant developments in our continuous delivery cycle. Check out our monthly video recaps on [YouTube](https://www.youtube.com/@zencoderai) to see these features in action. We're building Zencoder alongside you, and your [feedback](/get-started/community-support#share-your-feedback) shapes what comes next. ## Monthly Updates Zentester platform launch, new LLM call-based pricing model, /repo-info agent, image prompts in VS Code, and comprehensive platform improvements Zen Agents platform, MCP Library with 100+ servers, Marketplace launch, agent sharing, custom tools management, and change rollback features Bash/Shell tool available to coding agents, Requirements tool, MCP and client-search improvements, enhanced debugging capabilities, and UX refinements Coding Agent out of beta, "Coffee Mode", native integrations, Chrome extension for 20+ tools, MCP support, improved unit testing agent, and better docstrings generation # June 2025 Source: https://docs.zencoder.ai/changelog/june-2025 Zentester platform launch, new pricing model, /repo-info agent, image prompts in VS Code, and comprehensive platform improvements # Zentester Platform and Major Platform Enhancements June 2025 introduces significant new capabilities with the launch of our Zentester platform, a new pricing model, and powerful new features that enhance your development workflow. This release focuses on making Zencoder more intelligent, transparent, and capable of handling complex testing scenarios. Here's what we've been working on: ## New Features ### Zentester Platform with E2E Testing Agent We're excited to introduce the [Zentester platform](/features/e2e-testing) with our specialized E2E Testing Agent. This powerful new capability enables you to: * **Automated E2E test creation** is here and now you can use `/e2e-test` in chat to create comprehensive end-to-end tests in plain English * **Playwright integration** leverages Playwright for robust browser orchestration and automation * **Intelligent test recording** means that the agent can explore webpages, record user interactions, and automatically fix failing tests * **Cross-browser testing** lets you simulate user actions like scrolling, clicking, form filling, and snapshot verification across different browsers The E2E Testing Agent seamlessly integrates with your development workflow, automatically setting up Playwright if needed and generating tests that can be written for various testing frameworks. ### New Pricing Model: LLM Calls Instead of Token Usage We've completely redesigned our pricing structure to provide better transparency and predictability. Instead of pricing per token usage, we now price per LLM calls, offering: * **Enhanced visibility** by giving you detailed information about every prompt message and LLM interaction * **Predictable costs** and clear understanding of usage patterns with call-based pricing * **Better usage tracking** with Premium LLM call usage now displayed for all new pricing plans in your Zencoder account dashboard This represents a major shift in how we approach pricing transparency. Check out our updated [Pricing & Plans FAQ](/faq/plans) for complete details about the new pricing structure and plan options. ### /repo-info Agent Introducing the intelligent `/repo-info` agent that significantly enhances how Zencoder understands your project: * **Comprehensive project analysis** as the agent now collects detailed information about your repository architecture, dependencies, and build systems * **Context generation** through the creation of a comprehensive project file (`repo.md`) that serves as context for all subsequent agent requests * **Performance optimization** leading to a reduced number of LLM calls and speeding up agent responses by using pre-generated context This agent acts as a foundation for our [Coding agent](/features/coding-agent), providing it with deep project understanding from the start. ### Image Prompts in VS Code **Beta Feature**: Image prompts are currently in early beta stage. The agent can only use images that are part of your repository. We're actively working on adding the ability to use external images and copy-paste functionality in future updates. One of the most anticipated features is now available in VS Code. Users can now leverage visual context in their development workflow: * **Repository image support** lets you use images stored in your project repository as visual prompts * **Design-to-Code workflow** is as simple as uploading Figma screenshots or design mockups and prompting with: ``` Look at the image.png file and build me a component looking as close as possible to it ``` * **Visual context understanding** now enables agents to interpret visual designs and translate them into functional code We're actively working on bringing this capability to JetBrains IDEs and will continue improving image prompt functionality across all platforms. ## Documentation Updates We've significantly expanded our documentation to better support these new features: * **Updated pricing FAQ** brings comprehensive updates to our [Pricing & Plans FAQ](/faq/plans) explaining the new LLM call-based pricing model * **Known issues documentation** now comes with a new [Known Issues index page](/user-guides/known-issues/index) for better problem resolution * **JetBrains white screen fix** is now documented with the solution for the [white screen issue in JetBrains IDEs](/user-guides/known-issues/jetbrains-white-screen) * **E2E testing documentation** is now available with a comprehensive guide for our [E2E Testing Agent](/features/e2e-testing) ## Improvements ### Enhanced Agent Capabilities * **Code generation enhancement** means that our code generation in the code editor now uses the full Coding Agent for more intelligent and context-aware suggestions * **Bulk and single change reversion** lets our users now revert the latest changes done by agents in bulk or individually, providing better control over agent modifications * **Step limit management** means that we've added warning messages when agents reach maximum steps, with an option to continue the request ### MCP (Model Context Protocol) enhancements * **OAuth support** means that MCP servers now support OAuth authentication for secure third-party integrations * **SSE/HTTP streaming** allows remote MCP servers to support Server-Sent Events (SSE) and HTTP streaming without requiring OAuth * **Improved error messages** that we've added help provide better error handling and user feedback when installing MCP tools ### User Experience improvements * **Usage visibility** now shows the Premium LLM call usage clearly displayed for users on new pricing plans right after the agent is done with its work * **Interface refinements** bring general user experience improvements and interface polish across all platforms * **Stability enhancements** also introduce comprehensive improvements to platform stability and reliability * **One-click installation** lets you install official and third-party custom agents with a single click within the IDE ## Bug Fixes ### Shell Tool Improvements * **Stability fixes** have resolved freezing issues that occurred in certain scenarios * **Pagination handling** is now improved and the Shell tool properly avoids pagination in output for cleaner results * **UTF-8 encoding** improvements have fixed issues with UTF-8 with BOM encoding that weren't working properly in some cases ### Additional IDE fixes * **VS Code fixes**: * Fixed erroneous diff showing for empty files * Resolved cases where code completion wasn't displayed * Fixed rare cases where error screen was shown instead of chat * **JetBrains fixes**: * Performance improvements and fixes for IDE freezing * Enhanced diff review and single change revert functionality * Improved unsaved changes dialog for custom agents ### General platform UX fixes * **Apply button Fix** - Disabled Apply button for ghost messages to prevent confusion * **Error handling** - Improved error messages and user feedback across the platform ## Version History * 2.14.1 (June 30, 2025) * 2.14.0 (June 27, 2025) * 2.12.2 (June 19, 2025) * 2.10.0 (June 11, 2025) * 2.8.0 (June 5, 2025) * 2.6.1 (June 27, 2025) * 2.6.0 (June 23, 2025) * 2.5.0 (June 16, 2025) * 2.4.1 (June 11, 2025) * 2.4.0 (June 9, 2025) * 2.3.3 (June 5, 2025) * 2.3.2 (June 4, 2025) * 2.3.1 (June 3, 2025) * 2.3.0 (June 2, 2025) # March 2025 Source: https://docs.zencoder.ai/changelog/march-2025 A month of building and refinement with new integrations, Coffee Mode, and enhanced coding and testing capabilities # Coding Agent Out of Beta, Integrations and MCP Support Available March was a busy month for our team as we focused on making our AI coding assistant more intuitive, powerful, and integrated with your development workflow. Here's what we've been working on: ## New Integrations * **Jira Integration**: You can now mention Jira tickets directly in your chat requests, making it easy to fix or implement tickets without context switching. * **MCP Servers Configuration**: We've added seamless integration with Zencoder, opening the door to millions of community-built tools. This significant upgrade enables powerful new workflows across your development environment. * **Google Chrome Extension**: Our new [browser extension](https://chromewebstore.google.com/detail/zencoder/obkkfjgjhdkjnkleopbljciifoefhnbp) enables connection with 20+ different developer tools, making it easier to work with your favorite resources without leaving your workflow. This creates a seamless experience across IDEs and web-based tools. Explore all our integration options in the Integrations and MCP documentation. ## Introducing Coffee Mode ☕ Say goodbye to constant confirmations! Our new "Coffee Mode" lets your coding assistant work continuously without interrupting your flow. This productivity booster was one of our most requested features and is now available across all supported environments. ## Coding Agent Enhancements The Coding Agent is now out of beta and is generally available across all supported IDEs, bringing meaningful code quality improvements. Additionaly, we've significantly improved our coding agent's capabilities: * Better codebase understanding and more stable code generation * Intelligent prompting to use the coding agent when it's the better tool for the job * The `/fix` command now uses the coding agent instead of basic chat * Custom AI instructions are now supported Explore how the Coding Agent generates multi-file, multi-language code with deep codebase understanding. ## Unit Test Agent Improvements The unit tests agent got smarter: * Now checks for syntax and compile errors in generated code * Analyzes existing tests to avoid duplication * Overall improved generation quality * Thanks to the redesigned unit testing agent, you can now follow up in your chats more easily than before * Unit test agent now has access to Bash tools and can execute safe commands directly for you from the chat * The agent is aware of command outputs, making it easier for you to work more seamlessly Discover how the Unit Test Agent generates comprehensive test suites for your code. ## Custom Agents * You can now use custom agents in chat and see your custom agent name displayed in the chat interface. * We've improved the UI for defining custom agents, making it simpler and better integrated in the settings pane. Learn how to create specialized agents tailored to your specific development workflows. ## UX & Quality of Life Improvements Throughout March, we've been refining the user experience with UI, UX, and quality of life improvements — often overlooked but incredibly important changes that make your daily workflow smoother: * Single-click application for multi-file changes * Persistent coding agent state between chat sessions * New chat button & UI for easier creation from any screen * Ability to attach files from outside the project * Enhanced colors in various themes * Automatic saving of accepted changes to files * Fixed undo/redo behavior in chat input ## Docstrings Generation We've completely revamped our docstrings generation capability: * Simplified UI for a more intuitive experience * Dramatically improved speed and reliability * Better understanding of the code context that's being documented ## Documentation Overhaul We've made significant improvements to our documentation: * Completely recategorized content for easier navigation * Added a new [Concepts page](/get-started/basic-concepts) with a full product demo video * Added a new [Community, Feedback and Support hub](/get-started/community-support) for better user engagement * Started our changelog tracking (yes, you're reading it right now—how meta!) ## Version History * 1.13.0 (Mar 4) * 1.14.0 (Mar 6) * 1.15.0 (Mar 12) * 1.16.0 (Mar 12) * 1.17.0 / 1.18.0 (Mar 20) * 1.19.0 (Mar 25) * 1.16.0 (Mar 3) * 1.17.0 (Mar 11) * 1.18.0 (Mar 12) * 1.19.0 (Mar 20) * 1.20.0 (Mar 27) # May 2025 Source: https://docs.zencoder.ai/changelog/may-2025 Introducing Zen Agents platform with MCP Library and Marketplace for discovering and sharing AI agents and tools # Zen Agents Platform with MCP Library and Marketplace Launch May 2025 marks a significant milestone in Zencoder's evolution with the launch of our Zen Agents platform including our MCP Library and Marketplace. Releases from this month transform Zencoder from an individual productivity tool into a collaborative platform empowering entire development teams to create, share, and discover specialized AI agents and tools. Here's what we've been working on: ## New Features ### Zen Agents Platform Our [AI Agents](/features/ai-agents) (also known as Zen Agents) platform brings powerful customization and sharing capabilities to your development workflow: * **Agent Sharing** - Users can now share custom agents with all other users in their organization * **Custom Agent Registry** - We migrated the registry of custom agents to our backend, enabling synchronization across all your devices and IDEs * **Tool Configuration** - You can now easily configure and save the tools available to custom agents (Zencoder agents use all available tools by default) Explore how to create, customize, and share AI agents with your team ### MCP Library and Tools Management We've introduced a centralized way to manage and discover Model Context Protocol (MCP) servers and agent tools within Zencoder: * **Agent Tools Section** - Bringing together all tools used by agents in a dedicated UI for managing all tools * **MCP Library** - It's a built-in library currently featuring 100+ of the most useful MCP servers for engineers, ready to enable for your agents (plus you can add your own custom MCPs, as needed) * **Project-Level Installation** - MCP servers are now installed at the project level by default, simplifying configuration Discover how to browse, install, and manage MCP servers through our new Agent Tools interface ### Zen Agents Marketplace Going beyond the scope of the IDE - we've launched an open marketplace where you can discover and share specialized AI agents: * **Browse Specialized Agents** - Explore 20+ purpose-built agents covering the entire software development lifecycle * **Community Contributions** - Go ahead and submit your own agents to our [open-source repository](https://github.com/zencoderai/zenagents-library) * **Available in both IDEs and on the website** - Both IDEs and the website offer access to the same MCP and AI agents lists Discover a growing collection of specialized agents and MCPs to improve your development workflow ### Change Management Improvements * **Revert and Rollback Changes**: New ability to revert agent changes and review what has been changed - essentially giving you an undo function and more control over the agent output and higher granularity over agent behavior. ## Platform Improvements and Bug Fixes ### Shell and Bash Tool Improvements * Shell tool command output is now displayed in terminal in most cases * We've also fixed Shell Tool on WSL and resolved Powershell execution issues and stuck commands * Fixed compatibility issues with the Execute Shell Command tool in some terminals * Fixed an issue with agents entering an infinite cycle when executing bash tool commands ### Other Enhancements and Fixes * **UI Refinements** - Various UI improvements for a smoother experience * **Automatic Reconnection** - We now reconnect automatically if websocket connection is lost * **BYOK Support** - Added support for Bring Your Own Key (BYOK) Anthropic integration with specialized endpoints * **Android Studio Compatibility** - To play nice with Android Studio, we've fixed notification about the absence of JCEF on the latest Android Studio version ## Documentation Updates We've expanded our documentation to cover all the new features and improvements: * Updated [AI Agents](/features/ai-agents) page with comprehensive information about the Zen Agents platform, including creation, sharing, and management * Enhanced [Integrations and MCP](/features/integrations-and-mcp) documentation with details about the new MCP Library and Agent Tools interface * Updated [FAQ](/faq/general) with new information about fine-tuning options * Refreshed [Subscription](/get-started/subscription) page with some links and additional details ## Version History * 2.6.0 (May 29, 2025) * 2.4.0 (May 21, 2025) * 2.2.0 (May 14, 2025) * 2.0.0 (May 8, 2025) * 2.2.2 (May 26, 2025) * 2.2.1 (May 26, 2025) * 2.2.0 (May 22, 2025) * 2.1.1 (May 20, 2025) * 2.1.0 (May 18, 2025) * 2.0.0 and 2.0.1 (May 9, 2025) # General Questions Source: https://docs.zencoder.ai/faq/general Answers to common questions about Zencoder. Zencoder is a developer-focused AI tool that enhances the coding experience by providing intelligent code suggestions, contextual explanations, and seamless workflow integrations, enabling developers to achieve a state of flow in their projects. Zencoder stands out by: Yes, Zencoder prioritizes security by: Zencoder is designed for: Zencoder is compatible with: Absolutely. Zencoder's Repo Grokking feature provides deep contextual understanding, allowing it to offer precise suggestions and insights, even in extensive codebases. # Pricing & Plans Source: https://docs.zencoder.ai/faq/plans Frequently asked questions about Zencoder pricing plans, features, and usage limits. ## Plans & Pricing Basics | Plan | Price (per user/month) | Premium LLM Call Limits per day | Ideal For | | :----------- | :--------------------: | :-----------------------------: | :------------------------------------------- | | **Free** | **\$0** | **25** | Exploring Zencoder, no commitment | | **Starter** | **\$19** | **200** | Light individual usage & side projects | | **Core** | **\$49** | **550** | Professional devs, small to mid teams | | **Advanced** | **\$119** | **1500** | AI-first orgs, power users, heavy automation | The Free plan provides essential tools to get started. Starter includes a 2-week free trial. Core and Advanced plans include enterprise features like [Multi-Repo Indexing](/features/multi-repo), SSO and audit logs. A "Premium LLM Call" occurs each time the Agent interacts with the LLM (Large Language Model) while resolving your task. For example, if an Agent must use 5 different tools before giving you a final response, this process will consume a total of 6 Premium Calls: 5 for each tool interaction, plus 1 for the final response itself. This allows for fair pricing for complex tasks. Consider these two requests: * "Change label heading to Zencoder" * "Pull JIRA ticket DVL-0522, fix the bug, prepare PR, push to GitHub, and ping me on Slack when done" In both cases, it's one user message, but the latter would require more tool calls, and in general, assumes a more intelligent agent capable of following more complex instructions. The same logic applies to a simple edit vs a more complex one. **Why this pricing model?** * **Transparency**: You can see exactly how many LLM calls each task requires * **Control**: Optimize your agent instructions to be more efficient when needed * **Quality**: Our success is directly tied to delivering the most effective AI assistance possible, not cutting corners We will soon be offering a selection of Faster models (like Gemini Flash), which will consume a fraction of Premium LLM calls, and Advanced models (like Opus 4), which will consume multiple Premium LLM Calls, allowing you to further control the balance between quality and costs. **Why not price per token?** While token-based pricing would be simpler for us (pure pass-through), we chose LLM calls because they're easier for you to understand and manage. You can visually track tool usage and develop an intuition for which tasks require more calls. * **550 Premium LLM Calls** per user per 24h (resets 24h after the first agentic request on a given period) * **Courtesy mode** kicks in past the limit: responses slow down and may fall back to smaller models, but service doesn't cut off * **Seat-based**: each paid seat gets its own call bucket; buckets aren't shared or pooled ## Usage & Limits You can monitor your usage in two ways: **1) Real-time tracking after each message** After every agent usage, at the bottom of the message you can see how many premium LLM calls were consumed by the response, as well as how many are remaining for today and when the limit will be reset. Usage tracking shown after each agent message **2) Daily limit dashboard** You can view your daily limit usage by going to your user account at [https://auth.zencoder.ai](https://auth.zencoder.ai). Once logged in, navigate to the User Profile page where you'll find a "Limits" tab. You can also access it directly here. The daily usage limits dashboard in the auth.zencoder.ai portal is only available to users on the new plans (those who upgraded on June 19, 2025, or later). Users on legacy plans will not be able to see these usage metrics. Limits tab in user profile showing daily usage No. Basic autocomplete is currently unlimited across all plans. Only agent-driven tasks (chat, refactor, multi-file edits, etc.) count toward the daily Premium LLM Call total. If you reach your daily Premium Call limit, we will still provide courtesy access through a basic version of our Agent, although calls will be throttled and responses slowed down. For users who frequently rely on AI, we recommend upgrading to a plan better suited to your daily usage needs, or consider using our "Bring Your Own Key" option to remove daily usage limits altogether. Daily limits ensure you have consistent AI assistance every day of the month, not just the first week. With monthly limits, developers often burn through their entire allocation early in the sprint cycle, leaving them without access during critical deadlines. Daily limits also prevent high overage bills and make budget forecasting more predictable for teams. Additionally, we've built a buffer into our plans to accommodate usage spikes within the daily structure. No, unused Premium LLM calls do not roll over to the next day. Each day you get a fresh allocation of your plan's daily limits. This design ensures: * **Consistent daily access**: You're guaranteed your full allocation every day * **Predictable usage patterns**: Teams can plan their AI usage around daily cycles * **Fair resource distribution**: Prevents users from hoarding calls and ensures everyone gets consistent access * **Simplified billing**: No complex rollover calculations or expiration tracking Your daily limits reset 24 hours after your first agentic request of the current period, giving you a clean slate to work with fresh resources each day. Here are few tips to help you use AI more efficiently and avoid hitting daily limits prematurely: * **Use separate chats for different topics**: Agent might get confused by long chat history, causing quality issues and a need to use more tools to answer * **Use Basic Chat Agent for simple queries**: For questions that don't require a lot of code to be generated, and extensive codebase research (for example, quick questions about current file, or general software development knowledge questions) it's better to use Basic Chat Agent which is faster and more cost-effective * **Use custom agents for repetitive tasks**: Custom agents allow to specify custom prompts and tools to be reused, thus eliminating the need for agents to spend tool calls to investigate the details of what needs to be done * **Avoid "thank you" and other non-task-related messages**: While politeness is key in human interactions, it is not necessary, and even costly in interactions with agents. Please be aware that all messages are counted against the limits regardless of their content ## Account Management Once the trial ends, your account will automatically switch to the Free plan with limited features. You can choose to upgrade at any time to retain access to advanced tools and workflows. Upgrading is simple! Navigate to your account settings, select the plan you wish to upgrade to, and follow the on-screen instructions. Upgrading unlocks additional features and premium support. You can upgrade, downgrade, or cancel your plan from the self-serve customer portal (customer portal under profile). * **Upgrades** take effect instantly * **Downgrades** take effect on your next renewal * **Cancel** any time; you keep access until the end of the paid period For more information about our policies regarding cancellations and refunds, please see our detailed [Refund Policy](/faq/refund-policy). When you upgrade or change your Zencoder plan, the new features and permissions may not be immediately reflected in your IDE. Logging out and logging back in refreshes your authentication token and ensures your IDE recognizes your new plan's entitlements. In some cases, you may also need to restart your IDE for all changes to take effect. How to log out and log back in:

Visual Studio Code 1. Open the Command Palette (On macOS: Press `Shift + Command + P`, On Windows/Linux: Press `Shift + Ctrl + P`) 2. Type `Zencoder: Sign out` and select it VS Code Sign Out and Authentication Process 3. After signing out, you'll see large buttons to sign in - just follow the process JetBrains IDEs (IntelliJ, PyCharm, WebStorm, etc.) 1. Open Settings (On macOS: Press `Command + ,`, On Windows/Linux: Press `Ctrl + Alt + S` or go to `File > Settings`) 2. Navigate to `Tools > Zencoder` 3. Click the `Authenticate` button under "Launch authentication process" JetBrains Authentication Settings 4. Follow the prompts to sign out and sign back in
No. We understand that some of your users are more active than others. The reason we can provide this pricing to you is that AI usage varies day by day and user by user. We can average that, bringing you a convenient and predictable way to manage your AI budget.
## Advanced Options Yes. For most users, our pricing gives an easy and cost-effective way to plan and manage AI costs. Some users are comfortable spending \$200+/usr/mo directly on API providers as they feel they are getting 10x ROI on their use; in that case, BYOK is a good option. In the future, we could replace it with per-token pricing through the platform, if there's a demand for that. You pay Zencoder the seat fee (Starter/Core/Advanced). When you select the BYOK option and use your own OpenAI, Anthropic, or Gemini key, all LLM calls using that model will utilize your key. ## Legacy Plans Yes. You can switch to the current Starter tier on the Billing page. The seat price stays at \$19, but once you leave the legacy Business plan, you can't switch back. If you Move to Core, your price rises by +\$10, you keep SSO and Audit Logs. In addition, you will have visibility into LLM calls & usage. Move to Advanced (\$119) price rises, but you have 1500 daily LLM calls which is a much higher limit than the core and enterprise plan. Either way, once you leave the \$39 Enterprise plan, you can't revert. ## General Support Zencoder supports payment via: * All major credit cards * Apple Pay * Link * Cash App and Klarna For specific payment methods, wire transfers, or enterprise invoicing options, please [contact our support team](/get-started/community-support). Zencoder prioritizes security through: Zencoder supports integrations with popular IDEs, including Visual Studio Code, JetBrains IDEs, and Android Studio. We offer a comprehensive 14-day free trial that allows you to fully evaluate Zencoder before making any financial commitment. For detailed information about our refund policies, cancellations, and subscription management, please visit our [Refund Policy](/faq/refund-policy) page. # Privacy & Ethics Source: https://docs.zencoder.ai/faq/privacy Zencoder is committed to protecting your data and ensuring a secure and ethical AI experience. Zencoder employs robust security protocols to protect your code and data, ensuring compliance with industry standards and best practices. Zencoder is committed to ethical AI practices, including transparency, fairness, and accountability in its AI-driven coding assistance tools. Zencoder implements appropriate technical and organizational measures to safeguard your information against unauthorized access, disclosure, alteration, or destruction. Depending on your jurisdiction, you may have rights to: To report any concerns or incidents, please complete the [AI Adverse Impact Reporting Form](https://docs.google.com/forms/d/e/1FAIpQLScPgka564_ZDc3c30fL17VFuyRQmwUb95rxpe9bGjlb3kFdOA/viewform). This helps us assess, address, and respond to potential risks responsibly. # Product Features Source: https://docs.zencoder.ai/faq/product-features Zencoder offers a range of features designed to enhance your coding experience. Zencoder provides code generation, code improvement, unit test generation, docstring creation, and a codebase-aware chat feature to assist developers throughout the coding process. Zencoder's AI agents analyze your code to identify potential issues and suggest corrections, ensuring adherence to best practices and reducing errors. Yes, Zencoder integrates with various development tools and platforms to enhance your existing workflow. Zencoder utilizes embedded AI agents for code generation, unit testing, and docstring creation, providing precise suggestions by analyzing your code and integrating seamlessly into your workflow. Zencoder leverages a specialized proprietary AI pipeline for code generation, ensuring that the generated code is usable from the start. This proprietary AI pipeline achieves industry-leading compilation rates, making the development process more efficient. Zencoder automates unit test creation and execution, helping maintain high code quality and reducing the time developers spend on writing tests manually. Zencoder automates docstring generation, allowing developers to focus on high-value tasks and ensuring smooth knowledge transition between team members. Zencoder integrates with popular IDEs such as Visual Studio Code and JetBrains, including IntelliJ IDEA, PyCharm, and WebStorm. # Refund Policy Source: https://docs.zencoder.ai/faq/refund-policy Frequently asked questions about Zencoder refund policies, cancellations, and subscription management. ## Zencoder Refund Policy We believe in transparency and want you to feel confident about your investment in Zencoder. That's why we offer a generous 14-day free trial — giving you plenty of time to explore our AI agents, test them with your codebase, and see the productivity gains firsthand before making any financial commitment. Since we provide a full-featured 14-day trial to help you make an informed decision, we don't offer refunds for early cancellations. However, you'll continue to have full access to all your plan's features until your current billing period ends — so you can maximize the value of your investment. Downgrades take effect at the start of your next billing cycle, allowing you to fully utilize your current plan's capabilities until then. While we don't provide refunds for the difference, this approach ensures a smooth transition without service interruption. To maintain a safe and productive environment for all users, accounts suspended or terminated due to policy violations are not eligible for refunds. We're always here to help if you have questions about our usage policies. Absolutely! In the unlikely event we need to close your account for reasons unrelated to policy violations, we'll promptly refund you for any unused portion of your subscription. Your trust matters to us. Not at all! Our 14-day trial is completely free with no credit card required. You can explore all features, integrate with your tools, and experience the full power of Zencoder before deciding whether to upgrade or continue with our free plan. While we don't offer refunds for service interruptions, our team works around the clock to maintain 99.9% uptime and resolve any issues swiftly. We're committed to keeping your development workflow running smoothly. We don't currently offer subscription pausing, but many developers find our free plan perfect for maintaining access during breaks. When you're ready to dive back in, upgrading takes just a few clicks. Good news — you're protected from price increases during your current subscription term! Any pricing changes only affect future renewals, and we'll always notify you well in advance. We don't offer refunds based on usage. Whether you use the service a little or a lot, the subscription fee remains the same for the term. Our subscription model provides access to help you work at your own pace. Whether you're coding daily or tackling weekend projects, the value is there when you need it. That's why we encourage using the free trial to ensure Zencoder fits your workflow. # Agents Overview Source: https://docs.zencoder.ai/features/agents-overview Discover and access all AI agents available in Zencoder - from core development agents to custom AI assistants Zencoder provides a collection of AI agents designed to support your development workflow. These specialized agents can handle everything from answering questions about your code to generating comprehensive test suites and managing complex multi-file operations. ## Accessing Agents All agents are easily accessible through a unified interface: **Mac**: Press `Cmd + .` (Command + Period)\ **Windows/Linux**: Press `Ctrl + .` (Control + Period) Agent selector dropdown showing available agents Select from the dropdown menu showing both Core Agents and your custom AI Agents. The green section in the image contains Zencoder's default core agents, while the blue section shows your [custom AI agents](/features/ai-agents). You can also use the search bar at the top to quickly find specific agents. Agent dropdown showing core agents in green section and custom agents in blue section with search functionality Begin interacting with your selected agent immediately The agent selector shows all available agents in one convenient dropdown, with Core Agents at the top and your custom AI Agents below. ## Core Agents These specialized agents are built and maintained by Zencoder, each designed for specific development tasks and workflows. Your primary interface for asking questions about your codebase. Get explanations, seek guidance on best practices, and receive contextually relevant answers based on your project's structure and code. The primary development agent and workhorse of Zencoder. Generates, modifies, and refactors code across multiple files and languages with deep codebase understanding, validation, and error correction. Generates comprehensive test suites for individual components and functions. Creates appropriate mocks, assertions, and test cases tailored to your testing framework and coding patterns. Creates end-to-end tests for web applications using Playwright. Simulates user interactions, validates complete workflows, and ensures your application behaves correctly across different scenarios. Creates and maintains comprehensive project snapshots that improve all other agents' understanding. Analyzes your project structure, dependencies, and conventions to provide persistent context. ## AI Agents (Zen Agents) Beyond the core agents, Zencoder enables you to create and use custom AI agents tailored to your specific workflows and requirements. Create specialized agents for recurring tasks like code refactoring, documentation updates, or domain-specific workflows. These agents can be shared with your team and customized to match your development standards and practices. ## Typical Agent Workflow For new projects, start with the [Repo-Info Agent](/features/repo-info-agent) to establish comprehensive project context The [Q\&A Agent](/features/qa-agent) is perfect for understanding existing code, getting explanations, and planning your approach When you need to create or modify code across multiple files, the [Coding Agent](/features/coding-agent) provides the advanced capabilities needed for complex development tasks Use the [Unit Testing](/features/unit-testing) and [E2E Testing](/features/e2e-testing) agents to ensure your code is thoroughly tested ## Agent Relationships Understanding how agents work together can improve your development workflow: Running the Repo-Info Agent first provides comprehensive project context that improves the performance and accuracy of all other agents. It's recommended to run this agent when you first open a project. Use the Coding Agent to implement features, then leverage the Unit Testing and E2E Testing Agents to ensure comprehensive test coverage for your new code. The Q\&A Agent is ideal for asking questions about your codebase and getting advice while maintaining full understanding of your project context. Use it for general questions and planning - it won't make code changes but serves as an excellent starting point for understanding your current project. Each agent is designed to excel at specific tasks. Feel free to start new chats with different agents to work on specific tasks as your needs change throughout your development workflow. # AI Agents Source: https://docs.zencoder.ai/features/ai-agents Use prebuilt or create custom AI agents (we also call them Zen Agents) that can be called for selected code or entire files directly in your IDE ## Understanding AI Agents AI Agents in Zencoder are customizable, and sharable agents transforming how you code. Whether you're using our prebuilt agents or creating custom ones, these intelligent agents can be called for selected code or entire files directly in your IDE, helping you automate repetitive tasks, maintain coding standards, and improve your overall development efficiency. ## Navigating Agent Categories At the top of the AI Agents screen, you'll find tabs that help you filter and navigate between different types of agents: Shows all available agents Displays prebuilt agents created by Zencoder (with the number indicating how many are available) Shows your custom agents (with the number indicating how many you've created) ![AI Agents Selection](https://mintlify.s3.us-west-1.amazonaws.com/forgoodaiinc/images/ai-agents-selection.png) This categorization makes it easy to distinguish between default agents and your own custom creations. ## Creating and Managing AI Agents 1. Click the three dots menu (⋮) at the top right corner. 2. Select **Agents** 3. Click the **Add custom agent** button in the top right ![Create a Custom Agent](https://mintlify.s3.us-west-1.amazonaws.com/forgoodaiinc/images/ai-agents-make-custom-agent.png) When adding or editing an agent, you'll need to set these properties: **Basic Settings:** * **Name**: A descriptive name for your agent (required) * **Share**: Control who can see and use this agent * Personal: Visible only to you * Shared: Visible to your entire organization (all users within your organization can share agents with others) **Configuration:** * **Command/Alias**: The chat command to invoke your agent (required) * **Instructions**: Detailed prompt instructions that define your agent's behavior (required) * **Tools**: Enable specific tools your agent can access ![Configuring an Agent](https://mintlify.s3.us-west-1.amazonaws.com/forgoodaiinc/images/ai-agents-config.png) Define clear instructions for your agent that explain its purpose and how it should operate. Here's an example of instructions for a Repository Analyzer agent: ``` Consider the currently open project and repository. Understand its structure, mutual relations between files, folders, dependencies, and more. Please provide me with an in-depth structured understanding of this repo ``` Your instructions should be specific enough to guide the agent's behavior, but also flexible enough to work across different contexts. Agents can access various tools to enhance their capabilities: * git * File Search * Full Text Search * Web Search * Fetch Webpage Content * Semantic Code Search * File Editor * Execute Shell Command * Gather Requirements * ... Simply click on the tools your agent should have access to in the **Tools** section. ![Configuring Agent tools](https://mintlify.s3.us-west-1.amazonaws.com/forgoodaiinc/images/ai-agents-tools.png) If you're an admin, you can share agents with your organization: 1. When editing an agent, find the **Share** dropdown 2. Choose between: * **Personal**: Visible only to you * **Shared**: Visible to the whole organization ![Sharing Agent](https://mintlify.s3.us-west-1.amazonaws.com/forgoodaiinc/images/ai-agents-share.png) ## Using the Agents Marketplace The Zencoder Agents Marketplace is constantly growing! We're regularly adding new specialized agents to help with different aspects of your development workflow. Discover a growing collection of specialized agents to improve your development workflow ### Browsing the Marketplace The marketplace contains specialized agents for various development tasks, from bug prevention to documentation generation. Visit the marketplace to explore all available agents. ### Using Agents from the Marketplace Our Zencoder extension allows you to integrate powerful coding agents directly into your IDE. Follow these steps to browse and use agents from our Marketplace. Visit our Agents Marketplace to explore available agents. Browse through the list and select the one that best suits your development needs. Click on the agent to view its details, including: * Name * Description * Instructions * Tools required 1. Open the Zencoder extension in your IDE. 2. Click the three dots menu (⋮) at the top right corner. 3. Select **Agents** to open the custom agent creation page. 4. Click **Add Custom Agent**. * **Name**: Give your agent a unique name. * **Instructions**: Copy the instructions from the agent's detail page on the Marketplace and paste them into the "Instructions" field. * **Tools**: * Open the **Tools** section in the extension. * Navigate to **Manage Tools**. * Search for the tools listed in the agent's details and install them if needed. Tool configuration is case-by-case. Follow the guidance provided in the agent's description. There are no standardized patterns yet. * After configuring the agent with the name, command, instructions, and tools, save it. * You can now use this custom agent directly in your IDE for coding assistance. ### Contributing Your Own Agents We encourage all users to contribute to the Agents Marketplace! Whether you've created a specialized agent for a specific framework, language, or development task, sharing it helps grow our community of developers. Have you created a custom agent that others might find useful? Join our growing community of contributors! ## Using Agents in Your Workflow Once configured, you can use agents by: 1. Selecting code in your editor or focusing on an entire file 2. Invoking the agent using its command (e.g., `/unittests`, `/review`) 3. The agent will perform its specialized task on the selected code ## Related Agents and Capabilities Core agents and features that work alongside your custom AI agents: Quick questions and explanations for your custom agent development The primary development agent that your custom agents can complement Generate tests for code created by your custom agents Create comprehensive test scenarios for applications built with custom agents Provides project context that enhances all custom agents # Code Completion Source: https://docs.zencoder.ai/features/code-completion AI Code Completion with Zencoder Speed up your coding process with Zencoder’s AI-powered code completion feature. Zencoder intelligently predicts your next lines of code, offering context-aware suggestions that are precise and reliable. Open your IDE and ensure that Zencoder’s extension is installed and active. The code completion feature should be enabled by default. Begin coding as usual. Zencoder will automatically provide real-time suggestions based on your input and project context. Use the keyboard shortcut Tab or your mouse to accept the suggested code snippet. Refine as needed to align with your project’s requirements. ### Benefits of Code Completion with Zencoder * **Enhanced Productivity**: Write code faster with real-time, context-aware suggestions. * **Error Reduction**: Minimize typos and syntax errors with intelligent predictions. * **Improved Workflow**: Stay focused on your tasks without interrupting your flow to search for solutions. # Generate New Code Source: https://docs.zencoder.ai/features/code-generation AI code generation with Zencoder ## Overview Zencoder makes generating new code intuitive and fast. With Zencoder’s intelligent AI agents, you can create custom code snippets, automate code generation tasks, and receive contextual suggestions directly in your IDE. Whether developing from scratch or adding new features, Zencoder provides powerful assistance to help you code smarter and more efficiently.