cut the bs, gotta put examples

Author: Howon Lee

docs/guides/agent_chat.md

@@ -1,51 +1,63 @@
 # CodeRabbit Agentic Chat
 
-CodeRabbit Agentic Chat is a powerful assistant designed to streamline your development workflow. You can interact directly in pull request review comments and pull request comments to ask questions, have CodeRabbit rewrite code, or generate new code. This tool is built to enhance your productivity by integrating seamlessly into your existing development environment.
+CodeRabbit Chat is a powerful assistant designed to streamline your development workflow. You can interact directly in pull request review comments and pull request comments to ask questions, have CodeRabbit rewrite code, or generate new code. This tool is built to enhance your productivity by integrating seamlessly into your existing development environment.
 
 # Getting Started
 
-Agentic Chat is available exclusively for Pro customers. // some crap about sales process
+Chat is available exclusively for Pro customers. // some crap about sales process
 
 # Invocation
 
-To invoke CodeRabbit Agentic Chat, simply comment on a pull request using the handle @coderabbitai without any explicit command.
+To invoke CodeRabbit Chat, simply comment on a pull request using the handle @coderabbitai without any explicit command. The permission flow for the chat is the same as the deterministic CodeRabbit commands // some crap about commands
 
-Once activated, CodeRabbit processes your request and responds with the necessary assistance. This assistance may include generating scripts, updating issue tracking systems, offering detailed code suggestions, or even submitting an entire pull request, providing a streamlined and efficient workflow.
+Upon being invoked, CodeRabbit processes your request and responds with the necessary assistance. This assistance may entail generating scripts, updating issue tracking systems, offering detailed code suggestions, or even submitting an entire pull request.
 
 # Features
 
 ## Script Writing
 
-CodeRabbit can automatically generate scripts in various programming languages. It not only writes shell scripts but also selects the most appropriate scripting language and structure based on the specific query provided. This adaptive approach minimizes manual coding efforts by automating repetitive or standardized tasks, ultimately increasing efficiency and reducing development time.
+CodeRabbit Chat can automatically generate scripts in various programming languages (mainly shell scripting) to answer questions about your codebase. If incorrect or non-running or non-germane shell scripts are created, CodeRabbit goes and tries again, and the whole script iteration process will be displayed to you in an Analysis Flow.
+
+///////// example some crap
+///////// example some crap
 
 ## Web Search
 
-The assistant integrates real-time web search capabilities, enabling it to fetch up-to-date information to support its responses. By leveraging current data, CodeRabbit ensures that its answers are both accurate and contextually relevant. This feature assists in validating assumptions and discovering new methodologies, contributing to more informed and effective decision-making during development.
+CodeRabbit Chat integrates real-time web search capabilities, enabling it to fetch up-to-date information to support its responses. The CodeRabbit Chat response will indicate this if it decides to search the web to answer your query.
+
+// example
+// how to turn it off
 
 ## Jira and Linear Integration
 
-CodeRabbit offers seamless integration with popular project management tools such as Jira and Linear. This connection allows users to manage issue tracking and task updates directly from their development environment. By synchronizing pull request comments and code updates with these systems, the tool reduces manual data entry and helps maintain a smooth, uninterrupted workflow across the project lifecycle.
+CodeRabbit Chat is integrated with Jira and Linear.
+
+// link to authing it
 
 ## Learnings Integration
 
-The agent is designed to incorporate learnings and best practices into its workflows. Users can instruct CodeRabbit to remember specific pieces of information, which it then integrates into future tasks. This dynamic learning capability enables the tool to continuously refine code quality and improve overall efficiency by applying accumulated insights from previous interactions.
+CodeRabbit Chat can insert and delete Learnings from your Learnings database. Simply ask it to do so in plain natural language.
+
+////// example some crap
 
 ## Single Step Code Generation
 
-For rapid prototyping and quick fixes, CodeRabbit is capable of generating code snippets in a single step. This functionality is ideal for scenarios where developers need to copy and paste code quickly, reducing the likelihood of errors by automating repetitive tasks. The streamlined process saves time and allows developers to focus on more complex aspects of their projects.
+For rapid prototyping and quick fixes, CodeRabbit Chat is capable of generating code snippets in a single step.
+
+// example
 
 ## Multi-Step Agentic Flow with Planning
 
-When facing complex coding challenges, CodeRabbit supports multi-step agentic flows that involve detailed planning and execution. This sophisticated approach enables the tool to manage multi-stage tasks effectively, providing structured and iterative improvements to the code. Once the multi-step workflow is complete, CodeRabbit can automatically issue a pull request for the changes, ensuring that even intricate processes are handled efficiently.
+When facing complex coding challenges, CodeRabbit Chat supports multi-step agentic flows that involve detailed planning and execution. Once the multi-step workflow is complete, CodeRabbit can automatically issue a pull request for the changes, ensuring that even intricate processes are handled efficiently.
+
+// example
 
 ## Configuration Changes
 
 CodeRabbit is also capable of managing its own configuration settings upon request. This self-management allows the assistant to dynamically adapt to changing project requirements. By simplifying configuration updates, CodeRabbit helps users quickly adjust to new workflows or requirements without interrupting the development process.
 
-# Usage Guidelines
-
-When using CodeRabbit, it is important to provide clear and explicit instructions to ensure that the responses are accurate and relevant. Users are encouraged to leverage the integration with tools like Jira, Linear, and the web search functionality to maximize workflow efficiency. If the generated solution does not meet expectations, providing feedback will help the agent refine its output. Additionally, users should adhere to security best practices by avoiding the sharing of sensitive information during interactions.
+// example
 
-# Support and Contributions
+## Help
 
-For additional support, bug reports, or to contribute to the development of CodeRabbit Agentic Chat, users should refer to the project repository and follow the contribution guidelines provided. Active engagement with the community is encouraged, as it helps continuously improve the agent and its features. This collaborative approach ensures that CodeRabbit remains a valuable tool for individual developers and larger teams alike.
+Need help? Join our community on [Discord](https://discord.gg/coderabbit) or contact our support team.