Prompts #1
Merged
Prompts #1
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
- Updated `README.md` to include detailed usage instructions for custom prompts, including examples for `ralph.sh` and `ralph-once.sh`. - Modified `ralph.sh` and `ralph-once.sh` to enforce tool permission requirements when using the `--prompt` option, ensuring either `--allow-profile` or `--allow-tools` must be specified. - Added new prompt files in the `prompts` directory, providing structured guidance for various use cases, including WordPress plugin development and safe write-only operations. - Implemented a logging mechanism in `run-prompts.sh` to facilitate testing of prompts in isolated git worktrees, enhancing the testing process for prompt files.
- Updated README.md to clarify default prompt location and usage instructions. - Modified prompts/README.md to reflect changes in prompt file references. - Deleted prompts/default-agent.txt as it was no longer needed. - Revised prompts/safe-write-only.txt and prompts/wordpress-plugin-agent.txt to enhance clarity and instructions regarding PRD files and command execution. - Enhanced ralph-once.sh and ralph.sh scripts to support custom PRD file usage and improved error handling for prompt and PRD file readability. - Added new PRD JSON files for default-agent, safe-write-only, and wordpress-plugin-agent to define functional and UI test cases. - Created prompts/default.txt to serve as the new default prompt with comprehensive instructions for users. - Updated test/run-prompts.sh to include progress rendering and improved logging for prompt execution.
…ctionality and clarity - Set "passes" to true in prd.json for a specific test case. - Enhance progress.txt with additional implementation details and next steps. - Standardize prompt file references to clarify the source of truth. - Add error handling for unreadable progress files in ralph-once.sh and ralph.sh. - Consolidate PRD and progress file context for Copilot CLI to avoid multi-@ parsing issues. - Update run-prompts.sh to improve tool policy management and output capturing. - Introduce a comprehensive .gitignore to manage temporary files and artifacts.
… test cases to streamline project structure and improve maintainability.
…h-once.sh - Added sections for minimal invocation and parameters for both scripts. - Clarified the required positional argument and optional parameters. - Included notes on tool permission profiles and their implications. - Improved clarity on how prompts and PRD files are handled in the scripts.
- Refactor README.md to enhance organization and readability, including clearer section headings and streamlined content. - Revise the "Plans" section in plans/README.md to provide a more detailed explanation of the PRD format and best practices for creating work items. - Add examples and clarify the purpose of the `prd.json` file, emphasizing its role in guiding the AI agent. - Improve formatting for better visual clarity and user understanding.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This pull request introduces a major rewrite and expansion of documentation and configuration files for the Ralph Copilot CLI runner project. The changes focus on making the onboarding process clearer, improving usability for different workflows, and providing better guidance on PRD (Product Requirements Document) formats and prompt usage. Several new example PRD files and prompt files are added to support different agent permission profiles and use cases.
Documentation improvements:
README.md: Completely rewritten and reorganized for clarity. Adds a "Quick Start" section, detailed explanations of how Ralph works, configuration options, command reference, demo instructions, and project structure. Also clarifies model selection, permission profiles, and prompt usage.plans/README.md: Expanded to explain the PRD format, best practices, per-prompt PRDs, and file purposes. Provides clear guidance for customizing work items and using different PRD files with the runner.prompts/README.md: New file describing how to use sample prompt files and control tool permissions via script flags.PRD and prompt examples:
plans/prd-safe-write-only.json,plans/prd-wordpress-plugin-agent.json: New example PRD files for different agent permission profiles and workflows (e.g., safe write-only, WordPress plugin development). [1] [2]prompts/default.txt,prompts/safe-write-only.txt: New prompt files tailored for general and restricted agent behaviors, providing clear instructions for feature implementation and progress reporting. [1] [2]Project progress and usage:
progress.txt: Updated with a log entry describing the implementation of core message-sending behavior, setup of project scripts, and notes about restricted environment permissions.plans/prd.json: Marks the first work item as complete (passes: true) to reflect progress.Overall, these changes make the project easier to understand and extend for new users and contributors, while supporting more flexible agent workflows and permission models.