Added `gptel-pr-review` to `init-ai.el` for reviewing pull requests with specific prompts using gptel. Introduced `forge-pullreq-diff` in `init-git.el` to retrieve PR diffs as strings. These changes integrate AI capabilities for improved PR feedback and comment functionality.
Added a custom variable `gptel-ollama-context-window` to specify the number of characters for Ollama requests and overridden the `gptel--request-data` method to include this context window in the request data.
This commit introduces aimenu, a new package that generates imenu-like outlines using a language model. The new package includes functions for generating and handling outlines, managing cache, and interacting with the user to select headers.
Updated the gptel-directive for org-mode to provide detailed instructions on how to return Emacs Lisp, Python, and shell-script code blocks. Additionally, included several examples to illustrate the expected code block format and evaluation criteria.
This commit introduces two new directives into the gptel configuration: one for code review and another for prompt generation. The code review directive provides guidelines for giving feedback on code snippets, while the prompt generation directive outlines steps for creating effective prompts.
This change adds a new directive to the gptel configuration for creating detailed prompts for stable diffusion image generators. The directive specifies how to translate a simple natural language description into a more detailed and descriptive prompt to ensure high-quality image generation.
This commit introduces the gptel-chat-with-buffer function, allowing users to start a chat with the buffer content. It also updates keybindings, assigning "b" to the new function and moving gptel-select-backend to "B".
Implemented a gptel-org function to allow interacting with GPTel within an Org-Mode buffer, setting the appropriate system message for context. Updated init-org.el to load Emacs-Lisp and Python Babel languages and disabled org-confirm-babel-evaluate for smoother code execution.
This change introduces a new directive for generating shell commands to the gptel configuration. The directive specifies that the response should be plain text with no additional description or formatting.
Changed the model names from "llama3:latest" and "mistral:latest" to "llama3.1:latest" and "mistral-nemo:latest" respectively in the GPTel Ollama backend configuration.
Updated the guidelines for generating commit messages to avoid using the word 'enhance' and user experience references, and request plain text responses without syntax.
Reorganized backend definitions and selection process for gptel, introducing `gptel-backend-openai` and `gptel-backend-ollama` variables. Simplified backend selection logic and improved the `gptel-select-backend` function to dynamically prompt for models.
Enhanced the gptel configuration to support multiple backends (Ollama and OpenAI) and added a function to select the desired backend. Default backend is set to OpenAI.
