diff --git a/docs/features/api-configuration-profiles.mdx b/docs/features/api-configuration-profiles.mdx index e7cf5a5b..ed3d2cab 100644 --- a/docs/features/api-configuration-profiles.mdx +++ b/docs/features/api-configuration-profiles.mdx @@ -44,7 +44,7 @@ Configuration profiles can have their own: - [Temperature settings](/features/model-temperature) for controlling response randomness - Thinking budgets - Provider-specific settings -- [Diff editing configuration](/features/fast-edits) +- Diff editing configuration (see [`apply_diff`](/advanced-usage/available-tools/apply-diff)) - Rate limit settings Note that available settings vary by provider and model. Each provider offers different configuration options, and even within the same provider, different models may support different parameter ranges or features. @@ -143,5 +143,4 @@ API keys are stored securely in VSCode's Secret Storage and are never exposed in - Integrates with [local models](/advanced-usage/local-models) for offline work - Supports [temperature settings](/features/model-temperature) per mode - Supports per-profile rate limits (configured here) and general [usage tracking/cost info](/advanced-usage/rate-limits-costs) -- Supports [Diff/Fast Edits](/features/fast-edits) for tailored code editing behavior. - +- Supports diff-based editing (see [`apply_diff`](/advanced-usage/available-tools/apply-diff)). diff --git a/docs/features/experimental/experimental-features.md b/docs/features/experimental/experimental-features.md index 7b0a445c..f9072f3e 100644 --- a/docs/features/experimental/experimental-features.md +++ b/docs/features/experimental/experimental-features.md @@ -1,11 +1,10 @@ --- -description: 'Explore Roo Code''s experimental features including concurrent file edits and power steering. Enable advanced capabilities that are still under development.' +description: 'Explore Roo Code''s experimental features. Enable advanced capabilities that are still under development.' keywords: - experimental features - "Roo Code beta" - "advanced features" - "concurrent file edits" - - "power steering" - "feature flags" --- @@ -33,7 +32,6 @@ The following experimental features are currently available: - [Custom Tools](/features/experimental/custom-tools) - Define TypeScript/JavaScript tools that Roo can call like built-in tools - [Concurrent File Edits](/features/experimental/concurrent-file-edits) - Edit multiple files in a single operation -- [Power Steering](/features/experimental/power-steering) - Enhanced consistency in AI responses - [Background Editing](/features/experimental/background-editing) - Work uninterrupted while Roo edits files in the background - [Image Generation](/features/image-generation) - Generate images from text prompts and save them to your workspace - [Run Slash Command](/advanced-usage/available-tools/run-slash-command) - Execute predefined slash commands for templated instructions and workflow automation diff --git a/docs/features/experimental/power-steering.md b/docs/features/experimental/power-steering.md deleted file mode 100644 index 05d2d7cc..00000000 --- a/docs/features/experimental/power-steering.md +++ /dev/null @@ -1,70 +0,0 @@ ---- -sidebar_label: 'Power Steering' -description: 'Improve Roo Code''s response consistency with Power Steering. This experimental feature reinforces mode definitions and custom instructions for better adherence to assigned roles.' -keywords: - - power steering - - LLM consistency - - mode adherence - - custom instructions - - experimental feature - - token optimization - - role definition ---- -import Codicon from '@site/src/components/Codicon'; - -# Power Steering (Experimental Feature) - -The "Power Steering" experimental feature (`POWER_STEERING`) is designed to enhance the consistency of Roo Code's responses by more frequently reminding the underlying Large Language Model (LLM) about its current mode definition and any custom instructions. - ---- - -## How It Works - -When Power Steering is enabled, Roo Code constantly reinforces the LLM's understanding of its assigned role (e.g., "You are a helpful coding assistant") and any specific guidelines provided by the user (e.g., "Always provide code examples in Python"). - -This is achieved by explicitly including the `modeDetails.roleDefinition` and `modeDetails.customInstructions` within the information sent to the LLM with each interaction. - -**Goal:** -The primary goal is to ensure the LLM adheres more strictly to its defined persona and follows user-specific instructions more consistently. If you find Roo deviating from its role or overlooking custom rules, Power Steering can help maintain its focus. - -**Trade-off:** -These frequent reminders consume additional tokens in each message sent to the LLM. This means: -* Increased token usage per message. -* Potentially higher operational costs. -* The context window may be filled more quickly. - -It's a balance between stricter adherence to instructions and resource consumption. - -**Default Status:** Disabled. - ---- - -## Technical Details - -* **Experiment ID:** `powerSteering` -* **Mechanism:** - * The feature's status is checked by the `getEnvironmentDetails` function. - * If enabled, the current mode's `roleDefinition` and `customInstructions` are added to the details sent to the LLM. - * These details are wrapped in `` tags and become part of the context for each LLM interaction. -* **Impact:** By frequently including the role definition and custom instructions, the LLM is steered to generate responses more aligned with these parameters. - ---- - -## Enabling This Feature - -Power Steering is managed within the "Experimental Features" section of Roo Code's Advanced Settings. - -1. Open Roo Code settings ( icon in the top right corner). -2. Navigate to "Advanced Settings". -3. Locate the "Experimental Features" area. -4. Toggle the "Power Steering" option. -5. Save your changes. -Settings for Intelligent Context Condensation and Power Steering - -For general information on experimental features, see [Experimental Features Overview](/features/experimental/experimental-features). - ---- - -## Feedback - -Please report any issues or suggestions regarding this feature on the [Roo Code GitHub Issues page](https://github.com/RooCodeInc/Roo-Code/issues). Your feedback is crucial for improving Roo Code. \ No newline at end of file diff --git a/docs/features/fast-edits.md b/docs/features/fast-edits.md deleted file mode 100644 index 3c8d5843..00000000 --- a/docs/features/fast-edits.md +++ /dev/null @@ -1,54 +0,0 @@ ---- -description: Discover how Roo Code's Fast Edits feature uses diffs for faster, safer file modifications. Learn about match precision and configuration options. -keywords: - - fast edits - - diff editing - - file modifications - - apply diff - - match precision - - editing efficiency - - truncation prevention ---- - -# Diff/Fast Edits - -:::info Default Setting -Fast Edits (using the "Enable editing through diffs" setting) is enabled by default in Roo Code. You typically don't need to change these settings unless you encounter specific issues or want to experiment with different diff strategies. -::: - -Roo Code offers an advanced setting to change how it edits files, using diffs (differences) instead of rewriting entire files. Enabling this feature provides significant benefits. - -:::note Per-Provider Setting -Diff editing configuration is set per [API Configuration Profile](/features/api-configuration-profiles), allowing you to customize editing behavior for different providers and models. -::: - ---- - -## Enable Editing Through Diffs - -Open the Roo Code pane settings by clicking the gear icon . The `Providers` section will be visible. Select the specific [API Configuration Profile](/features/api-configuration-profiles) you want to configure. - -When **Enable editing through diffs** is checked: - - Roo Code settings showing Enable editing through diffs -1. **Faster File Editing**: Roo modifies files more quickly by applying only the necessary changes. -2. **Prevents Truncated Writes**: The system automatically detects and rejects attempts by the AI to write incomplete file content, which can happen with large files or complex instructions. This helps prevent corrupted files. - -:::note Disabling Fast Edits -If you uncheck **Enable editing through diffs**, Roo will revert to writing the entire file content for every edit using the [`write_to_file`](/advanced-usage/available-tools/write-to-file) tool, instead of applying targeted changes with [`apply_diff`](/advanced-usage/available-tools/apply-diff). This full-write approach is generally slower for modifying existing files and leads to higher token usage. -::: - ---- - -## Match Precision - -This slider controls how closely the code sections identified by the AI must match the actual code in your file before a change is applied. - - Roo Code settings showing Enable editing through diffs checkbox and Match precision slider - -* **100% (Default)**: Requires an exact match. This is the safest option, minimizing the risk of incorrect changes. -* **Lower Values (80%-99%)**: Allows for "fuzzy" matching. Roo can apply changes even if the code section has minor differences from what the AI expected. This can be useful if the file has been slightly modified, but **increases the risk** of applying changes in the wrong place. - -**Use values below 100% with extreme caution.** Lower precision might be necessary occasionally, but always review the proposed changes carefully. - -Internally, this setting adjusts a `fuzzyMatchThreshold` used with algorithms like Levenshtein distance to compare code similarity. \ No newline at end of file diff --git a/docs/features/index.md b/docs/features/index.md index 91b96bd7..5946cfb7 100644 --- a/docs/features/index.md +++ b/docs/features/index.md @@ -19,7 +19,6 @@ Discover the powerful features that make Roo Code your ultimate AI-powered codin ## Core Features ### Code Editing & Navigation -- [**Fast Edits**](/features/fast-edits) - Lightning-fast code modifications with intelligent diff application - [**Concurrent File Reads**](/features/concurrent-file-reads) - Read multiple files simultaneously for better context - [**Code Actions**](/features/code-actions) - Quick fixes and refactoring suggestions - [**Diagnostics Integration**](/features/diagnostics-integration) - Real-time error detection and resolution @@ -60,7 +59,6 @@ Push the boundaries with cutting-edge capabilities: - [**Custom Tools**](/features/experimental/custom-tools) - Define TypeScript/JavaScript tools that Roo can call - [**Concurrent File Edits**](/features/experimental/concurrent-file-edits) - Edit multiple files in a single operation -- [**Power Steering**](/features/experimental/power-steering) - Enhanced consistency in AI responses - [**More Experimental Features**](/features/experimental/experimental-features) - Explore features under development ## Additional Resources @@ -71,4 +69,4 @@ Push the boundaries with cutting-edge capabilities: --- -Ready to explore? Start with our [Getting Started Guide](/getting-started/installing) or dive into any feature that interests you! \ No newline at end of file +Ready to explore? Start with our [Getting Started Guide](/getting-started/installing) or dive into any feature that interests you! diff --git a/docs/features/intelligent-context-condensing.mdx b/docs/features/intelligent-context-condensing.mdx index b8d83f93..c36f8ceb 100644 --- a/docs/features/intelligent-context-condensing.mdx +++ b/docs/features/intelligent-context-condensing.mdx @@ -40,12 +40,11 @@ As your conversation with Roo Code grows, it might approach the context window l 1. **Summarizing:** Using an AI model, it condenses earlier parts of the conversation. 2. **Retaining Essentials:** The goal is to reduce the overall token count while keeping the key information from the summarized messages. 3. **Maintaining Flow:** This allows the AI to have a more coherent understanding of the entire conversation, even very long ones. -4. **Preserving First Message:** The first message containing slash commands or initial context is always preserved during condensing, ensuring critical initial instructions aren't lost. +4. **Slash command continuity:** Slash commands included in the first message are preserved across condensations. **Important Considerations:** * **Summarization Impact:** While original messages are preserved if you use [Checkpoints](/features/checkpoints) to rewind, the summarized version is what's used in ongoing LLM calls to keep the context manageable. * **Cost:** The AI call to perform the summarization incurs a cost. This cost is included in the context condensing metrics displayed in the UI. -* **First Message Protection:** The initial message with slash commands or setup instructions will never be condensed. --- @@ -54,45 +53,41 @@ As your conversation with Roo Code grows, it might approach the context window l Intelligent Context Condensing is **enabled by default** and offers several configuration options: 1. Open Roo Code settings ( icon in the top right corner of the Roo Code panel). -2. Navigate to the "Context" settings section for context-related options, or the "Prompt" settings section for the custom prompt. +2. Navigate to the **Context Management** settings section. 3. Configure the available options: - **Automatically trigger intelligent context condensing**: Enabled by default, this controls whether condensing happens automatically (found in "Context" settings) - **Threshold to trigger intelligent context condensing**: A percentage slider (default 100%) that determines when condensing activates based on context window usage (found in "Context" settings) - - **API Configuration for Context Condensing**: Choose which API configuration to use for condensing operations (defaults to your current active configuration) (found in "Context" settings) - - **Custom Context Condensing Prompt**: Customize the system prompt used for context condensing operations (found in "Prompt" settings) + - **Custom Context Condensing Prompt**: Customize the prompt Roo uses when condensing (found in "Context Management" settings) + - **Condensing model**: Condensing uses your active conversation provider/model. Settings for Intelligent Context Condensing -*Intelligent Context Condensing configuration options: automatic triggering toggle, threshold slider, API configuration selection, and custom prompt customization.* +*Intelligent Context Condensing configuration options: automatic triggering toggle, threshold slider, and custom prompt customization.* --- ## Controlling and Understanding Context Condensing Roo Code provides several ways to control and understand the Intelligent Context Condensing feature: -### Controlling Context Condensing +#### Controlling context condensing * **Automatic Threshold:** The threshold slider in "Context" settings allows you to define a percentage (e.g., 80%) of context window usage. Roo Code will attempt to condense the context automatically when the conversation reaches this level of capacity. -* **API Configuration:** Select which API configuration to use for context condensing operations. This allows you to use a different provider or model specifically for condensing if desired. -* **Custom Prompts:** Modify the system prompt used for condensing to better suit your workflow or to emphasize certain aspects of conversation summarization. +* **Custom Prompt:** Modify the prompt used for condensing to better suit your workflow or emphasize what should be preserved. * **Manual Trigger:** A **Condense Context** button is available at the top of the task, positioned to the right of the context bar. This allows you to initiate the context condensing process at any time. Manual Condense Context button in expanded task view *The Manual Condense Context button (highlighted with a yellow arrow) is easily accessible for manual control.* -### Understanding Context Condensing Activity +#### Understanding context condensing activity * **Context Condensing Metrics:** When context condensing occurs, Roo Code displays: * The context token counts before and after context condensing. * The cost associated with the context condensing AI call. * An expandable summary detailing what was condensed (this information is part of the `ContextCondenseRow` component visible in the chat history). -Context condensed message in chat -*After context condensing, a message indicates the context has been condensed, showing token changes and cost.* +A completed context condense row showing token counts and cost +*Why this view matters: it’s your audit trail for what changed, how much it cost, and what Roo will carry forward.* * **Visual Indicators:** * A progress indicator ("Condensing context...") is shown in the chat interface while context condensing is active. -Condensing context progress indicator in chat -*The "Condensing context..." indicator appears in the chat during the process.* - * The task header also displays the current context condensing status. * The `ContextWindowProgress` bar offers a visual representation of token distribution, including current usage, space reserved for the AI's output, available space, and raw token numbers. * **Interface Clarity:** The "Condense Context" button includes a tooltip explaining its function, available in all supported languages. @@ -101,15 +96,15 @@ Roo Code provides several ways to control and understand the Intelligent Context ## Tips for Effective Context Condensing -### Customizing the Context Condensing Prompt +#### Customizing the context condensing prompt You can customize the context reduction prompt to better suit your specific domain or use case. This is particularly useful if you find that the default condensing process loses important information specific to your workflow. To customize the prompt: 1. Go to Roo Code settings ( icon) -2. Navigate to the "Prompt" settings section -3. Find the "Custom Context Condensing Prompt" in the pulldown -4. Enter your custom prompt that instructs the AI on how to preserve specific types of information +2. Open **Context Management** +3. Find the **Custom Context Condensing Prompt** editor +4. Enter your custom prompt that instructs Roo on what must be preserved For example, if you're working on a complex debugging session, you might add instructions like: - "Always preserve error messages and stack traces in full" @@ -124,19 +119,19 @@ This customization ensures that the context condensing process retains the infor When Roo Code encounters context window limit errors, it now automatically recovers to keep your work flowing: -### How Error Recovery Works +#### How error recovery works 1. **Error Detection**: Roo Code detects context window errors from multiple providers (OpenAI, Anthropic, Cerebras, and others) 2. **Automatic Truncation**: The system automatically reduces the context by 25% -3. **Retry Mechanism**: After truncation, Roo Code retries your request (up to 3 attempts) -4. **Seamless Continuation**: Your work continues without manual intervention +3. **Retry Mechanism**: After truncation, Roo Code retries your request (up to the built-in retry limit) +4. **Continuation**: Roo retries without manual intervention This automatic recovery ensures that: - You don't lose work due to context limit errors - Long conversations can continue smoothly - The system intelligently manages context without requiring manual restarts -### When Recovery Triggers +#### When recovery triggers The automatic recovery activates when: - The API returns a context window exceeded error @@ -149,7 +144,7 @@ This feature works alongside Intelligent Context Condensing to provide multiple ## Technical Implementation -### Token Counting +#### Token counting Roo Code uses a sophisticated token counting system that: - Employs native token counting endpoints when available (e.g., Anthropic's API) - Falls back to tiktoken estimation if API calls fail @@ -158,14 +153,22 @@ Roo Code uses a sophisticated token counting system that: - Image content: Uses a conservative estimate of 300 tokens per image - System prompts: Includes additional overhead for structural elements -### Context Window Management +#### Context window management - By default, 30% of the context window is reserved (20% for model output and 10% as a safety buffer), leaving 70% available for conversation history. - This reservation can be overridden by model-specific settings - The system automatically calculates available space while maintaining this reservation -### Error Handling Strategy -- **Multi-Provider Support**: Recognizes context errors from OpenAI, Anthropic, Cerebras, and other providers -- **Progressive Reduction**: Reduces context by 25% on each retry attempt -- **Retry Limits**: Maximum of 3 automatic retry attempts before requiring manual intervention -- **State Preservation**: Maintains conversation state during recovery attempts +#### Error handling strategy + +#### Why you can’t use a different model/provider for condensing (and why that’s good) + +Condensing always uses your active conversation provider/model. + +Using a different model to condense can degrade summary quality when the history includes tool calls, tool results, or other structured content. Keeping condensing on the same model/provider avoids “translation” errors between different tool/format expectations. + +#### How the custom condense prompt is stored (advanced) + +Your custom condense prompt is stored using the same support-prompt override mechanism as other prompt templates: `customSupportPrompts.CONDENSE`. +- If you had a legacy `customCondensingPrompt`, Roo migrates it into this location automatically. +- **Reset** clears the override so Roo falls back to the built-in default condense prompt. diff --git a/docs/features/settings-management.md b/docs/features/settings-management.md index 47d57349..b0ad2a78 100644 --- a/docs/features/settings-management.md +++ b/docs/features/settings-management.md @@ -45,7 +45,10 @@ Clicking the **Import** button allows you to load settings from a previously exp 2. A file open dialog appears. Select the `roo-code-settings.json` file (or similarly named file) you want to import. 3. Roo Code reads the file, validates its contents against the expected schema, and applies the settings. * **Merging:** Importing settings **merges** the configurations. It adds new API profiles and updates existing ones and global settings based on the file content. It does **not** delete configurations present in your current setup but missing from the imported file. -* **Validation:** Only valid settings matching the internal schema can be imported, preventing configuration errors. A success notification appears upon completion. +* **Validation:** Import validates the file, but it can still succeed with warnings. + + - If **some** API profiles reference a provider that no longer exists (or is otherwise invalid), Roo imports the rest and reports warnings. + - Import fails only if **all** profiles are invalid. --- @@ -184,4 +187,4 @@ With both disabled, these sections are omitted, reducing token usage when you do - Disabled: Thinking blocks are expanded by default. - Notes: - Applies across conversations globally. - - Text is localized; labels may differ by language. \ No newline at end of file + - Text is localized; labels may differ by language. diff --git a/docs/update-notes/index.md b/docs/update-notes/index.md index 929b8af0..ab7c3a4a 100644 --- a/docs/update-notes/index.md +++ b/docs/update-notes/index.md @@ -21,12 +21,16 @@ If you want to live on the edge and try things out before it's released, we have --- +### Version 3.43 + +* [3.43.0](/update-notes/v3.43.0) (2026-01-23) + +--- ### Version 3.42 * [3.42.0](/update-notes/v3.42.0) (2026-01-22) --- - ### Version 3.41 * [3.41.2](/update-notes/v3.41.2) (2026-01-16) diff --git a/docs/update-notes/v3.11.0.md b/docs/update-notes/v3.11.0.md index b4edeb8a..81fe1b27 100644 --- a/docs/update-notes/v3.11.0.md +++ b/docs/update-notes/v3.11.0.md @@ -21,7 +21,7 @@ This release introduces faster diff editing, project-level MCP configuration, AP Roo Code's default editing mechanism now uses a multi-block diff strategy for the [`apply_diff`](/advanced-usage/available-tools/apply-diff) tool. This is significantly faster, especially when applying multiple changes at once, as it modifies only the necessary lines instead of rewriting the entire file. This leads to quicker edits and helps prevent issues like truncated writes on large files. -Learn more about [Fast Edits](/features/fast-edits). +Learn more about [`apply_diff`](/advanced-usage/available-tools/apply-diff). --- @@ -85,4 +85,4 @@ Learn more about [Import/Export/Reset Settings](/features/settings-management) a * **Tool-Calling Logic Refactor:** Significant internal refactoring. (Thanks diarmidmackenzie, bramburn, KJ7LNW, and others!) * Displayed info about partial reads in the chat row. * Linked to the settings page from the auto-approve toolbar. -* Linked to provider docs from the API options. \ No newline at end of file +* Linked to provider docs from the API options. diff --git a/docs/update-notes/v3.11.md b/docs/update-notes/v3.11.md index df214eef..02110242 100644 --- a/docs/update-notes/v3.11.md +++ b/docs/update-notes/v3.11.md @@ -21,7 +21,7 @@ This update focuses on performance enhancements, improved provider integration, Roo Code's default editing mechanism, which uses diffs via the [`apply_diff`](/advanced-usage/available-tools/apply-diff) tool, is now significantly faster, especially when applying multiple changes at once. This approach modifies only the necessary lines instead of rewriting the entire file, leading to quicker edits and helping prevent issues like truncated writes on large files. This reduces waiting time and improves the flow of iterative development. -Learn more about [Fast Edits](/features/fast-edits). +Learn more about [`apply_diff`](/advanced-usage/available-tools/apply-diff). --- @@ -115,4 +115,4 @@ Learn more about [Interacting with Suggestions](/features/suggested-responses#in * The `list_code_definition_names` tool now supports individual files. (Thanks KJ7LNW!) * File @-mentions now preserve language-specific characters. (Thanks aheizi!) * Fixed Gemini command escaping. -* Fixed switching profiles to ensure only the selected profile is switched (thanks feifei325!). \ No newline at end of file +* Fixed switching profiles to ensure only the selected profile is switched (thanks feifei325!). diff --git a/docs/update-notes/v3.43.0.mdx b/docs/update-notes/v3.43.0.mdx new file mode 100644 index 00000000..dda1d811 --- /dev/null +++ b/docs/update-notes/v3.43.0.mdx @@ -0,0 +1,44 @@ +--- +description: This release improves Intelligent Context Condensation, updates providers, and fixes export and settings issues. +keywords: + - roo code 3.43.0 + - new features + - bug fixes +image: /img/v3.43.0/v3.43.0.png +--- + +# Roo Code 3.43.0 Release Notes (2026-01-23) + +This release updates Intelligent Context Condensation, removes deprecated settings, and fixes export and settings issues. + +Roo Code v3.43.0 Release + +## Intelligent Context Condensation v2 + +Intelligent Context Condensation runs when the conversation is near the model’s context limit. It summarizes earlier messages instead of dropping them. After a condense, Roo continues from a single summary, not a mix of summary plus a long tail of older messages. If your task starts with a slash command, Roo preserves those slash-command-driven directives across condenses. Roo is less likely to break tool-heavy chats during a condense, which reduces failed requests and missing tool results. + +Settings changes: the Condense prompt editor is now under **Context Management** and Reset clears your override. Condensing uses the active conversation model/provider. There is no separate model/provider selector for condensing. + +([#10873](https://github.com/RooCodeInc/Roo-Code/pull/10873), [#10920](https://github.com/RooCodeInc/Roo-Code/pull/10920), [#10909](https://github.com/RooCodeInc/Roo-Code/pull/10909), [#10927](https://github.com/RooCodeInc/Roo-Code/pull/10927), [#10931](https://github.com/RooCodeInc/Roo-Code/pull/10931), [#10876](https://github.com/RooCodeInc/Roo-Code/pull/10876), [#10881](https://github.com/RooCodeInc/Roo-Code/pull/10881), [#10901](https://github.com/RooCodeInc/Roo-Code/pull/10901)) + +## QOL Improvements + +* Removes the unused “Enable concurrent file edits” experimental toggle to reduce settings clutter. ([#10925](https://github.com/RooCodeInc/Roo-Code/pull/10925)) +* Removes the experimental **Power Steering** setting (a deprecated experiment that no longer improves results). ([#10926](https://github.com/RooCodeInc/Roo-Code/pull/10926)) +* Removes obsolete diff/match-precision provider settings that no longer affect behavior. ([#10298](https://github.com/RooCodeInc/Roo-Code/pull/10298)) +* Adds a `pnpm install:vsix:nightly` command to make installing nightly VSIX builds easier. ([#10912](https://github.com/RooCodeInc/Roo-Code/pull/10912)) + +## Bug Fixes + +* Fixes an issue where MCP config files saved via the UI could be rewritten as a single minified line. Files are now pretty-printed. (thanks Michaelzag!) ([#9864](https://github.com/RooCodeInc/Roo-Code/pull/9864)) +* Fixes an issue where exporting tasks to Markdown could include `[Unexpected content type: thoughtSignature]` lines for some models. Exports are now clean. (thanks rossdonald!) ([#10932](https://github.com/RooCodeInc/Roo-Code/pull/10932)) +* Fixes an issue where the **Model** section could appear twice in the OpenAI Codex provider settings. ([#10930](https://github.com/RooCodeInc/Roo-Code/pull/10930)) + +## Misc Improvements + +* Removes legacy XML tool-calling code paths that are no longer used, reducing maintenance surface area. ([#10929](https://github.com/RooCodeInc/Roo-Code/pull/10929)) + +## Provider Updates + +* Updates Z.AI models with new variants and pricing metadata. (thanks ErdemGKSL!) ([#10860](https://github.com/RooCodeInc/Roo-Code/pull/10860)) +* Corrects Gemini 3 pricing for Flash and Pro models to match published pricing. (thanks rossdonald!) ([#10487](https://github.com/RooCodeInc/Roo-Code/pull/10487)) diff --git a/docusaurus.config.ts b/docusaurus.config.ts index a6e86c7c..251ed17a 100644 --- a/docusaurus.config.ts +++ b/docusaurus.config.ts @@ -176,6 +176,10 @@ const config: Config = { to: '/features/intelligent-context-condensing', from: ['/features/experimental/intelligent-context-condensing', '/features/experimental/intelligent-context-condensation'], }, + { + to: '/features/experimental/experimental-features', + from: ['/features/experimental/power-steering'], + }, { to: '/advanced-usage/footgun-prompting', from: ['/features/footgun-prompting'], @@ -309,6 +313,12 @@ const config: Config = { to: '/', from: ['/providers/claude-code'], }, + + // Redirect removed Fast Edits feature page + { + to: '/', + from: ['/features/fast-edits'], + }, ], }, ], diff --git a/sidebars.ts b/sidebars.ts index 3bd860da..f2866e41 100644 --- a/sidebars.ts +++ b/sidebars.ts @@ -21,24 +21,23 @@ const sidebars: SidebarsConfig = { { label: 'Features', type: 'category', - items: [ - 'features/api-configuration-profiles', - 'features/auto-approving-actions', - 'features/boomerang-tasks', - 'features/browser-use', - 'features/checkpoints', - 'features/code-actions', - 'features/codebase-indexing', - 'features/custom-instructions', - 'features/diagnostics-integration', - 'features/custom-modes', - 'features/fast-edits', - 'features/enhance-prompt', - 'features/settings-management', - 'features/intelligent-context-condensing', - 'features/keyboard-shortcuts', - 'features/marketplace', - 'features/message-queueing', + items: [ + 'features/api-configuration-profiles', + 'features/auto-approving-actions', + 'features/boomerang-tasks', + 'features/browser-use', + 'features/checkpoints', + 'features/code-actions', + 'features/codebase-indexing', + 'features/custom-instructions', + 'features/diagnostics-integration', + 'features/custom-modes', + 'features/enhance-prompt', + 'features/settings-management', + 'features/intelligent-context-condensing', + 'features/keyboard-shortcuts', + 'features/marketplace', + 'features/message-queueing', 'features/model-temperature', 'features/concurrent-file-reads', 'features/rooignore', @@ -73,7 +72,6 @@ const sidebars: SidebarsConfig = { 'features/experimental/experimental-features', 'features/experimental/custom-tools', 'features/experimental/concurrent-file-edits', - 'features/experimental/power-steering', 'features/experimental/background-editing', 'features/image-generation', 'advanced-usage/available-tools/run-slash-command', @@ -173,6 +171,11 @@ const sidebars: SidebarsConfig = { label: 'Extension Release Notes', items: [ 'update-notes/index', + { + type: 'category', + label: '3.43', + items: [{type: 'doc', id: 'update-notes/v3.43.0', label: '3.43.0'}], + }, { type: 'category', label: '3.42', diff --git a/static/img/v3.43.0/v3.43.0.png b/static/img/v3.43.0/v3.43.0.png new file mode 100644 index 00000000..b38ad925 Binary files /dev/null and b/static/img/v3.43.0/v3.43.0.png differ