Execute persistent task loops with retries, verification, and debugging escalation.
npx skills add https://github.com/AGMO-Inc/agmo-everywhere --skill ralph-agmo-inc
Please help me install this Agent Skill. Skill: ralph Source: https://github.com/AGMO-Inc/agmo-everywhere/tree/main/skills/ralph Command: npx skills add https://github.com/AGMO-Inc/agmo-everywhere --skill ralph-agmo-inc
This Skill provides a persistent completion loop that keeps executing work until all tasks are verified complete, automatically dispatching tasks, collecting failure context, and escalating to debugging when needed.
Start the Ralph skill and tell it to continue until all tasks are verified complete.
High-intent search queries and answers about installing and using this skill.
Persistent task execution automation continuously processes work items in a structured loop until verification passes. It dispatches tasks to appropriate executors, captures failure context, and retries up to three times before escalating to debugging if issues persist.
A verification loop processes work items and runs verification hooks after each step. When failures occur, it captures context and retries up to three times with preserved history. If all retries fail, it triggers a debugging escalation workflow to diagnose root causes.
Automated task dispatch with category routing directs work items to appropriate executors based on their category within a persistent loop. You provide a structured task list, and the loop continuously processes items, routing each to the correct executor until none remain.
Yes, the verification loop can pause when users interrupt or unrecoverable errors occur. It reports progress after each step, preserving the current state and failure history so execution can resume or be diagnosed through debugging escalation.
Managing retries and debugging escalation in task orchestration requires a structured loop with verification hooks. The loop retries failed tasks up to three times with preserved context, then triggers a debugging workflow to resolve root causes before final verification and exit.
Yes, persistent execution loops require a structured task list, verification hooks to confirm completion, and optional debugging guidance to manage failures safely. These elements ensure the loop can dispatch, verify, retry, and escalate without manual intervention.
stevengonsalvezOrchestrate research, planning, implementation, and validation phases with persistent state.
doravidanResume RALPH execution from the last checkpoint by loading intervention state.
marcusOrchestrate concurrent cursor-agent processes with timeouts and JSONL logging.
CodingHeaderCoordinate LLM agents to work in parallel on independent software-engineering tasks, aggregating their results and monitoring progress along the way.
