Documentation v2.0.0 #9
pranshugupta54
started this conversation in
Tips
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Traycer: AI-Powered Coding, Refined
What's Traycer?
Traycer is an AI-powered coding assistant that transforms how you handle coding tasks. With its Tasks feature, Traycer helps you define objectives, generate actionable plans, and implement changes seamlessly. Whether you need to fix bugs, refactor code, or create new features, Traycer analyzes your input and automates the process while keeping you in control. Alongside tasks, Traycer offers real-time, context-aware insights, identifying issues, optimizing performance, and improving code quality—all without disrupting your workflow.
Table of Content
Important Links
Features
Tasks
The Tasks feature allows you to define high-level objectives, and Traycer works autonomously to analyze and modify your code to accomplish them. Whether it's fixing bugs, creating new features, or optimizing existing code, you can rely on Traycer to handle complex coding tasks seamlessly.
task-demo-720p.mov
User Query
Enter your task in the input box (e.g., "Refactor the login module") and use @mention or the + icon to specify files or folders (e.g.,
@auth.js
,@utils/
). Once done, click the"Generate Plan"
button to proceed.Plan Specification
Traycer generates a plan with steps for your task. You can edit the plan directly, add or remove files, or include reference files for context. Click the
"Generate Changes"
button when ready.Code Changes
Traycer displays the code changes for each file in the plan in a structured view. You can click on any file to expand its details and review the changes. Each file includes options to Show Diff, allowing you to compare the proposed changes with the original code, and Apply, enabling you to integrate the changes into your codebase instantly. Selecting a file also opens a dedicated chat interface, where you can discuss, refine, or further customize the changes before applying them.
File-Specific Chat
Traycer provides a dedicated chat interface for each file in the task, enabling focused collaboration and refinement. Once you select a file, you can view the proposed changes and engage with Traycer to discuss, clarify, or request additional modifications. The chat ensures a streamlined workflow by offering the following capabilities:
Show Diff
for comparing changes andApply
to integrate updates seamlessly into your codebase.This file-specific chat enhances collaboration and ensures the proposed changes align perfectly with your goals.
Task History
Traycer maintains a log of your completed and ongoing tasks in the Task History panel. Each entry displays the task description and its current status, such as
"Planning"
,"Plan Generated"
etc. You can revisit any task to view its details or pick up where you left off, ensuring a seamless workflow and easy access to past objectives.Review
The Review feature allows you to analyze your code for potential issues, enhancements, and best practices. Traycer provides detailed suggestions to improve code quality, including bug fixes, performance optimizations, and adherence to coding standards. You can trigger reviews manually or let Traycer automatically analyze changes in real-time, ensuring your code is always at its best.
Basic_Usage.mp4
Triggering Analysis: Manual vs. Automated
Traycer provides flexibility in how you trigger code analysis:
Traycer continuously tracks your activities and performs analysis in real-time. Analysis is triggered automatically by events such as saving your work.
This allows users to initiate the analysis at their convenience through various methods, providing flexibility and control.
Analyze File
but is useful for re-evaluating the entire file without evaluating what has already been analyzed in a previous analysis.Switching Between Manual and Automated Analysis
Configure your preferences to switch between manual and automated analysis modes.
Different ways to switch between Manual and Automated Analysis
Using Command Palette: -
Ctrl+Shift+P
ORCmd+Shift+P
on Mac).Traycer: Enable Auto Analysis
ORTraycer: Disable Auto Analysis
.Extension Settings Page:
Go to the Traycer extension settings page in VS Code.
Change the "Traycer: Execution Mode" setting to "Manual" or "Auto".
execution-mode-settings.mp4
Manually Triggered Analysis
This allows users to have full control over when the analysis runs. Users can trigger the analysis process manually, making it useful for reviewing and refining code at specific points in the development workflow.
Different ways to trigger Analysis manually
Editor Toolbar: Click the Traycer icon on the top editor toolbar to trigger analysis on the current file.
Right-click Menu: Right-click anywhere in the code file and select Traycer >
Analyze File
,Re-Analyze File
,Analyze Changes in File
, orAnalyze Changes in Workspace
.Command Palette: Run
Traycer: Analyze File
,Traycer: Re-Analyze File
,Traycer: Analyze Changes in File
, orTraycer: Analyze Changes in Workspace
from the command palette by usingCtrl + Shift + P
orCmd + Shift + P
.Checking Analysis Status
To check the status of your analysis, hover over the Traycer icon. It will display the count of running analyses, such as "Traycer in-flight analysis: X". If the count is 0, it means there are no analyses currently running.
analysis-status.mp4
Comments & Replies
Traycer adds comments to your code, suggesting improvements and highlighting potential issues such as bugs, performance, security, and clarity. These comments provide actionable insights to help you refine your code, maintain high standards, and address problems early in the development process. Additionally, you can reply to comments to engage in a conversation with Traycer, allowing you to discuss and resolve issues directly within the context of each comment.
(A) Navigator
To navigate and manage comments in Traycer, follow these steps:
bugs
,performance
,security
, andclarity
.(B) Traycer Comment
The Comment Detailed View in Traycer provides an in-depth look at each comment, offering context and actionable insights to help improve your code. Here are some key features:
Code Snippet Display with Diff: The comment includes a relevant code snippet showing the differences between your current code and the suggested changes from Traycer AI. This helps you understand the exact context of the comment and the proposed improvements.
Reply and Discuss: You can engage in a conversation with Traycer by replying to comments. This allows you to discuss the issues directly within the context of each comment, making it easier to resolve problems and refine your code.
Quick Fix: The Wand icon allows you to directly apply the suggested fix, streamlining the process of addressing issues. If a comment does not have a suggested fix, the Wand icon will not be present.
Comment Deletion: The Bin icon allows you to delete the entire comment along with its conversation.
Feedback Options: You can give a thumbs up or thumbs down to provide feedback on the comment. This helps improve the accuracy and usefulness of future comments.
Actionable Insights: Each comment suggests improvements or highlights potential issues such as bugs, performance, security, and clarity. The detailed view provides a clear explanation of the issue and how to address it.
Learnings
Learnings enable Traycer to capture user preferences and code-relevant insights from your interactions. When you reply to comments, Traycer analyzes your responses to identify preferences or learnings, saving them to improve future analyses and recommendations.
How It Works
Managing Learnings
Accessing Learnings: Click on the Learnings icon in the menu to view your saved learnings.
See the list of all the learnings you've accumulated:
Deleting Learnings: Click the Bin icon next to the learning you wish to delete.
Adding Learnings Manually: You can manually add new learnings by clicking the Plus icon. Specify preferences or insights directly without waiting for Traycer to capture them in your replies.
Privacy Control: Your privacy is important. Traycer provides a setting in the Platform where you can opt out of the Learnings feature anytime. Opting out will delete all existing learnings. Note that the learning feature is on by default.
Analysis History
Keeps a record of all past analyses performed, allowing you to review and track changes over time. Each entry includes key details such as the trigger type, state of completion, duration, and statistical breakdowns of the analysis. You can monitor the number of successfully analyzed code blocks, failed analyses, skipped sections, and other relevant metrics, helping you understand the effectiveness and progress of your code improvements.
To access the analysis history, follow these steps:
Preferences
Customize your Traycer settings to fit your development workflow and preferences.
Modifying Settings
To modify any setting, follow these steps:
Via Extension Settings Page
Ctrl + ,
(Cmd + ,
on Mac).Via Command Palette
Ctrl + Shift + P
(Cmd + Shift + P
on Mac).Via Side Panel Preferences
Execution Mode
Specifies how Traycer triggers analysis of files.
Auto: (Default) Traycer continuously tracks your activities and performs analysis in real-time as you write and modify your code. Analysis is triggered automatically by events such as saving your work.
Manual: Users have full control over when the analysis runs. You can trigger the analysis process manually at specific points in your development workflow, either for the entire file or only for recent changes.
Comment Indicator
Specifies how comments are displayed in the editor. You can choose from the following options:
Comments Highlighting
Specifies how comments are highlighted in the editor. Highlighting is useful for quickly identifying code lines with comments. You can enable or disable highlighting and customize the color to fit your preferences. The default setting is disabled.
Output Level
Specifies how much (if any) output will be sent to the Traycer output channel. You can configure the level of detail for the output, ranging from minimal to detailed logs. The default level is set to "info".
Feedback Forum
We value your input and encourage you to share your experiences and suggestions with us. Connect with us and the community in the following ways:
Discord: Join our community here
X (Twitter): Follow us @TraycerAI
GitHub: Encounter a bug or have a feature request? File an issue on our community repository.
Terms of Service and Privacy Policy
By using Traycer, you agree to our Terms of Service and Privacy Policy.
Beta Was this translation helpful? Give feedback.
All reactions