Changelog

Stay up-to-date with the latest improvements and fixes in PURECODE AI.

v5.0.7

August 15, 2025

Added:

  • Tool Permission Manager : centralized tool manager where users can select which tools to allow or deny. Includes global options for “Allow All” and “Ask Every Time.” The tool permission control is located in the chat area next to the MCP icon in Agent Mode. Clicking it opens the permission popup.

  • Summary for create and update files in Agent Mode : when users request file creation or updates, they now receive a short code summary explaining the code for convenience. In the same code snippet, this summary will be added at the bottom.

Fixed:

  • Agent Mode – Stream Interruption Fix : fixed an issue where a restart option appeared during code streaming, causing interruptions. This no longer occurs, ensuring a smoother experience.

Improved:

  • UI/UX and overall stability.

v5.0.6

August 8, 2025

Improved:

  • Improved functionality and overall system stability

v5.0.4

July 31, 2025

Added:

  • Added support for universal standard MCP config

  • Added fallback mechanism and handled wrong arguments format in read files tool call

Fixed:

  • Error boundary issue on extension sidebar when stopping tool flow execution.

Updated:

  • Gemini model terminal command run tool and terminal instance selection.

  • UI/UX and overall stability.

  • Exact Search tool call argument handling for malformed inputs.

v5.0.3

July 24, 2025

Added:

  • Session as Context Provider: Users can now provide full context of past conversations to PureCode AI. They can use the "@" mention to access "session" provider, view all previous chat sessions, or search a particular session using initial messages after using "@" mention.

  • Toggle & Refresh for MCP Section: Users can now disable specific MCPs by toggling them off, and the changes will be updated in a few seconds ( <4 sec ) . Additionally, after editing the JSON, users can use the refresh option to get the latest MCP details without needing to manually reload the window.

  • Accessible MCP from Chat Input box: When in agent mode, users will see an MCP icon directly in the chat area next to the model dropdown. Clicking this icon will open the MCP panel allowing users to add, remove and manage their MCP servers.

Fixed:

  • Malformed JSON in Create/Update File Tool call: We have resolved the issue with malformed JSON when creating or updating files in agent mode.

  • 400 Bad Request in Agent Flow: We have fixed the 400 bad request error that occurred when using the create/update file tool call in agent flow.

Improved:

  • Checkpoint Consistency: We have addressed an edge case to ensure consistent checkpoint creation for version control operations.

  • Agent Mode as Default: Agent mode is now set as the default mode. Users can switch to chat mode if needed in chat input box.

v5.0.2

July 17, 2025

Added:

  • Stable MCP Support: Introduced stable MCP (Model Context Protocol) support, enhancing reliability when connecting to MCP servers.

  • Gemini Model 2.5 Pro MCP Integration: Added support for MCP integration with Gemini 2.5 Pro, enabling advanced use cases and streamlined model interaction.

Fixed:

  • Planning Steps UI: Resolved edge case UI issue affecting the planning steps interface for a smoother and clearer experience.

  • Create/Update ToolCalls Handling: Fixed a bug where toolcalls would error out if interrupted during generation. Also resolved an issue where stopping a generation would wipe out the toolcalls response.

  • Gemini Agent 500 Internal Sever Error Fix: Resolved an issue where users received a 500 Internal Server Error when making a request in Gemini agent mode with MCP configured. The root cause was identified and corrected to ensure stable request handling.

  • Resolved Tool Name Conflict Between MCP and Agent: Resolved an issue where the model could pick tool names from both the MCP side and the agent side simultaneously, causing conflicts. Now, the tool name is selected from only one source at a time, preventing collisions and ensuring consistent behavior.

Improved:

  • UI/UX & Stability: Improved interface, usability, and stability across key extension functionalities in the PureCode AI VS Code extension.

  • Prompting for Gemini Agent: Enhanced the prompting system for the Gemini agent, improving response accuracy.

v5.0.1

July 15, 2025

Fixed:

  • File Access Error (EACCESS/EPERM): Fixed an issue where the model failed to read or edit files due to restricted or read-only directories. The fix improves permission handling and avoids invalid paths.

v5.0.0

July 11, 2025

Added:

  • MCP (Model Context Protocol) Integration Support: The PureCode AI Extension now supports integration with multiple MCP servers. This enables seamless connection to local binaries, Docker containers, or remote HTTP services with a simple JSON configuration format.

  • Create/Update Files Tool in Agent Mode: Users can now create and update files within agent mode.

  • Feedback Model for Issues, Suggestions, and Feature Requests: Users can provide feedback, report issues, or send feature requests by disliking a response. After disliking, they can select the "Help Us Improve" option, which opens a feedback view where users can choose the category type (e.g., issue, feature request, suggestion) and submit their message. The PureCode team will review and respond promptly.

  • Checkpoint Toggle Option: A new toggle in the settings panel lets users choose whether to enable or disable the visibility of checkpoints.

Improved:

  • Response Streaming: The streaming flow has been enhanced to provide a smoother user experience when receiving responses.

  • Prompting Improvements: Various improvements have been made to the prompting system to increase the accuracy and relevance of responses.

Fixed:

  • Scrolling Issue During Streaming: The UI scrolling issue during streaming has been resolved. Users can now view content properly, even when the code is expanded.

v4.1.0

July 05, 2025

Added:

  • New Payment Popup Notifications: We have introduced new payment popups within the Extension GUI. Users will now receive context-specific notifications directly in the interface. These popups will alert users about various conditions, such as account cancellation, past-due payments, credit usage, free account limits reached, and overdue accounts. This enhancement ensures that users are always informed about their account status without leaving the VS Code environment.

  • Restore Point Text Addition: A new informational text has been added to the restore point section, informing users that database changes and package installations cannot be reverted. This update helps users understand the limitations of the restore point functionality.

Fixed:

  • Agent Tool Call ID Mismatch Fix for 400 Error: The issue where users could not use tool calls with different models in the same session has been resolved. The flow will now be smooth, allowing users to switch between models seamlessly within a session.

  • Account Mismatch Fix for Token Sign-in: If users sign in with a different account and then paste a new token, the correct account details will be fetched and displayed in the profile section without needing to refresh the window. Additionally, if users navigate from the /components route to VS Code with another account, the token will be logged, and web-selected generation will be done directly within VS Code.

Improved:

  • Case Sensitivity Indexing Issue: Users will no longer face indexing issues when making case-sensitive modifications to their repo folder. It will now auto-index without any issues and pass the same context items as it did before changing the repo name with case sensitivity.

  • Improved Message Count Display: We have enhanced the message count display in the right section panel. Users will now see proper messages and a progress bar when they've used all chat messages.

  • Terminal Tool Call Enhancement: In Agentic flow users will now be able to see the model being used terminal tool call, instead of just being given terminal commands in the terminal code block.

v4.0.9

July 02, 2025

Added:

  • New Settings Panel: A comprehensive settings panel has been added, providing users with detailed options on the right side. This includes the Profile section, MCP section [ coming soon ] , a new Feedback section where users can request new features, report technical issues, and submit bug reports, as well as a Global & Project Chat Instructions view and a Features view.

    Users can now access the settings panel from the extension header, located right next to the profile option.

  • Accessible Docs Page: Users can view all PureCode AI features and documentation, including the changelog and more. The Docs page is accessible from the book icon in the extension header, which links directly to the official documentation site.

Fixed:

  • Checkpoints Issue: The issue where checkpoints wouldn’t appear after reopening the VS Code extension has been resolved. Users can now see the checkpoints even after closing and reopening VS Code.

  • Hold Shift Issue: Earlier, when dragging and dropping images, there was a display issue. This has now been fixed, and users can freely drag and drop images within the chat area, but this requires holding the Shift key as expected.

  • Generating State Mismatch: Users will no longer see the "generating" status after the code has already been generated. This edge case is now handled.

  • Reusing Previous Inputs: Users can now update previous chat inputs by modifying the providers and existing uploaded images, and can re-submit the request from there.

  • Agent-Initiated Tool Calls Are Credit-Free: Tool calls made in Agent Mode will no longer be counted towards the user’s quota or credits.

Improved:

  • Gemini Tool Call UI: The Gemini tool call UI has been fixed, and agent mode is now supported for the Gemini model. Tool calls now happen just like other models, enhancing user experience.

  • Default Gemini model : Now free users will get the gemini 2.5 pro as their default model for both chat & agent mode. They will get the option to switch to any model by selecting from the dropdown.

  • New Model Dropdown Design: The model dropdown has been updated to display newly added models and indicate the default model, providing better clarity.

  • Hiding Line Numbers: Line numbers in context items will now be hidden if no chunks are available, ensuring a cleaner interface.

  • Context Length Update: The context length for all models has been updated.

  • UI & UX Stability Improvements: UI & UX stability improvements have been made for smoother performance and a more reliable experience.

v4.0.8

June 27, 2025

Fixed:

  • Default profile selection for generation process : Now, if the personal account is not available, the system will fallback to using the team account to ensure team users can still make requests without any interruptions.

v4.0.7

June 17, 2025

Added:

  • New tool call in Agent mode — Exact search tool call for enhanced precision in searching.

  • Model prompting and client implementation to understand the OS terminal status, enabling it to provide related terminal code.

  • Ability for the model to continue terminal commands in the same terminal or open a new terminal in Agent mode.

  • Saving user team selection, model selection, and mode selection in local storage to retain the last selected options when opening a new window.

Fixed:

  • Image Upload issue resolved for better user experience.

  • Fixed issues in the "Read Files" tool call, ensuring proper file reading functionality.

  • Addressed the issue of empty and malformed commands in the terminal tool call, improving stability in Agent Mode.

  • "Accept All" and "Reject All" options now appear above chat area only after the model provides the full response, addressing the issue where they appeared prematurely.

Improved:

  • UI improvements, including enhancements to the Terminal tool call, Checkpoint UI, and the entire chat area, improving layout, flow, and overall usability.

v4.0.6

June 04, 2025

Added :

  • Dedicated Diff view to allow users to easily compare changes side-by-side in GUI section. Also users can seamlessly switch between the "Generated" code view and the new "Diff" view at any time after the initial generation is completed.

  • Replaced the 'Send' and 'Codebase' buttons with a single 'Send' button. Introduced a 'Use Codebase' checkbox (enabled by default), located next to the advanced indexing option below the chat area. When enabled, the context will be passed with the user query; if unchecked, the query will be sent without the context.

  • New scroll-to-bottom option added to help users quickly reach the bottom of the chat area. The button appears above the chat area in the center.

  • Floating edit option for recent input sections, enabling users to re-initiate requests directly from their previous input.

  • Custom client messages for better handling of status codes and responses.

Improved :

  • Code snippet height adjustments while streaming, ensuring better readability and a smoother user experience.

  • UI enhancements for the Planning Section design, providing a more intuitive layout.

  • Updated the tool call UI, including the Read Codebase tool call UI.

Fixed:

  • We've updated the model dropdown to clearly show which models support chat and which don’t support agent tasks.

  • Fixed token and config refetching issues when using multiple windows of the extension instance.

  • Resolved UI wipeout issue with @Mentions in the chat area when users write a prompt and mentions provider in the middle, along with an extra space issue.

  • Fixed the issue where users weren’t receiving the full response inside the snippet when requesting documentation. Now, the entire response from the model is properly included inside the snippet, ensuring nothing is lost.

v4.0.3

May 21, 2025

Added:

  • New tool calls in agent mode for reading problems and terminal issues, expanding the capabilities of agent mode.

  • New onboarding flow to guide users in discovering the full potential of PureCode AI extension features.

  • Custom model prompt tuning to handle edge cases for each model.

  • Back-and-forth chat with users. It enables interactive, humanized conversations for iterative coding support.

Improved:

  • Checkpoint improvements for agent mode to enhance stability.

  • Stability and UI/UX improvements for a smoother user experience.

  • Complete refactoring of prompt architecture.

  • New prompt for all models.

  • Updated tool call with all functionality.

  • Preventing model from hallucinating and generating irrelevant content.

Fixed:

  • User message edge cases being handled (greeting messages, faulty messages, etc.).

  • Planning steps fixed for Gemini model.

v4.0.2

May 10, 2025

Added:

  • Support for Checkpoint and Restore File Changes: This feature enables users to restore project files to specific snapshots, offering an easy way to roll back changes.

  • Agents for Codebase Context & Running Terminal Commands: Now you can manage the codebase context more efficiently and run terminal commands directly for streamlined workflows.

  • read_codebase Toolcall Implemented Using Provider \ A new toolcall method read_codebase has been introduced, implemented through a provider, offering enhanced access to the codebase for better management and context retrieval.

Improved:

  • Trimming Content of Terminal Provider to 150 Lines: To keep things clean and efficient, the terminal provider will now limit its output to 150 lines. This helps prevent too much clutter and ensures better performance without overwhelming you with excessive data.

  • Preventing the Calculation of Advanced Indexing on Queries for Agent Mode: When working in agent mode, we've stopped the system from calculating advanced indexing during queries. This change improves speed and ensures the process is more efficient, without unnecessary tasks slowing things down.

  • Preventing Context Passing in Agent Mode: Context won't be automatically passed along in agent mode anymore. This makes sure each operation is more focused, avoiding any interference from previous contexts and keeping things precise.

  • Files Visible in Provider Once Created or Added to Codebase: Now, as soon as you create or add files to your codebase, they’ll be visible in the provider. No need to wait or refresh; you’ll have immediate access to your files when you need them.

  • More Dynamic Actions and Management for Generated Codes: The Accept All and Reject All options now appear above the Continue Generating button, even when responses are interrupted. Users can accept/reject all at once and continue the generation process, allowing them to manage further responses as they come. This applies to all assistant responses with code blocks for easier code management.

Fixed:

  • UI/UX Stability and Improvements: Removed gradient loaders and replaced them with a more straightforward "Generating" status view, including an option to stop the process. This update improves visual clarity and access during task generation.

  • Image Input Copy/Paste Issue Fixed: Resolved issues where image inputs were being pasted twice in the chat area and where no response was being received after pasting and hitting send.

v 4.0.1

May 01, 2025

  • Switched default model to Gemini 2.5 pro preview

  • UI/Ux Improvements

v 4.0.0

April 26, 2025

New UI Feature: Smooth Version Navigation

  • New Interactive Feature: Users can now smoothly navigate between different chat versions by hovering over horizontal dots in the chat area, making it easier to switch and view previous states of the conversation.

New Model Support: Gemini 2.5 Pro, Claude 3.7 and GPT-4.1

  • We've integrated support for three advanced AI models—GPT-4.1, Claude 3.7, and Gemini 2.5 Pro into our platform. These additions provide users with access to cutting-edge AI capabilities, offering enhanced performance, improved accuracy, and greater versatility across various tasks.

v 3.0.9

April 17, 2025

Auth:

  • Removed the auth reducer from persistent storage.

  • Added an initial check in AuthWrapper from secretstorage to verify token availability. If the token is not present, it will attempt to fetch it from localStorage. If not available, the token will be removed from the Redux state.

UI/UX Update:

  • Updated the design and interaction of the "Continue Generation" button.

Bug Fix:

  • Fixed an issue where the cursor pointer would jump while editing the planning step and prompt.

  • Resolved the issue with the missing advanced indexing loader during generation.

Indexing Improvements and Bug Fixes:

  • Fixed the reminder logic to notify users about advanced indexing based on their preferences stored in the database.

  • Resolved an issue where normal indexing was not being computed unless advanced indexing was enabled.

  • Users are now prevented from enabling advanced indexing while normal indexing is in progress. A toast notification is shown in such cases.

  • All indexing operations are now stopped on user sign-out and will automatically restart after sign-in.

  • Fixed inconsistencies in indexing status updates, particularly during Git operations and other edge cases such as RepoMapInd.

PreviousFrequently Asked Questions

Last updated