Convert PRDs into prd.json and execute user stories with Ralph.
npx skills add https://github.com/gannonh/kata-agents --skill ralph-gannonh
Please help me install this Agent Skill. Skill: ralph Source: https://github.com/gannonh/kata-agents/tree/main/skills/ralph Command: npx skills add https://github.com/gannonh/kata-agents --skill ralph-gannonh
Ralph reduces the manual effort of translating PRDs into a machine-readable plan and orchestrating automated execution of user stories via the Ralph agent.
To start Ralph on a PRD, place your PRD or prd.json in the working directory and run the Ralph script: ./scripts/ralph.sh. Optional arguments let you choose the tool and max iterations, for example: ./scripts/ralph.sh --tool claude 15.
High-intent search queries and answers about installing and using this skill.
Autonomous execution handles sequential user stories by spawning a fresh AI instance per iteration to avoid memory leakage. Ralph runs these stories from prd.json, iterating through each until completion or until the maximum iteration limit is reached.
Autonomous execution handles sequential user stories by spawning a fresh AI instance per iteration to avoid memory leakage. Ralph runs these stories from prd.json, iterating through each until completion or until the maximum iteration limit is reached.
Branch archival automatically preserves results when the feature branch name changes during autonomous execution. Ralph triggers this archival process to maintain execution history and prevent data loss across different feature branches.
Branch archival automatically preserves results when the feature branch name changes during autonomous execution. Ralph triggers this archival process to maintain execution history and prevent data loss across different feature branches.
Ralph spawns a fresh AI instance per iteration during autonomous execution to avoid memory leakage across user stories. This ensures each story runs with a clean state, iterating sequentially until all stories are complete or maximum iterations are reached.
Ralph spawns a fresh AI instance per iteration during autonomous execution to avoid memory leakage across user stories. This ensures each story runs with a clean state, iterating sequentially until all stories are complete or maximum iterations are reached.
crsieblerConvert markdown or text PRDs into Ralph's prd.json schema.
doravidanMerge RALPH worktree changes into the main branch with optional squash.
CElyousfiConverts Markdown files to editable ASTN JSON and vice versa with formatting preserved.
sixtycat2000-ctrlConvert markdown or plain text PRDs into Ralph's prd.json format.
