Turn any /goal into the next concrete action. Make a shareable link for a human, or hand an agent a Ralph loop with fresh-state reading, bounded execution, verification, and handoff.
---
name: goal-ralph-loop-figure-out-x
description: Complete "/goal figure out x" with one bounded, verified Ralph loop.
goal: "figure out x"
goal_page: https://goalfigureoutx.com/goal/figure-out-x
source: https://goalfigureoutx.com/skills/goal-ralph-loop.md
homepage: https://goalfigureoutx.com
---
# Goal Ralph Loop: figure out x
## Trigger
Use this skill when an agent receives this exact goal:
```text
/goal figure out x
```
## Goal Page
https://goalfigureoutx.com/goal/figure-out-x
## Workflow
1. Read fresh state for "figure out x": inspect current instructions, files, tests, task trackers, and recent verification output.
2. Aim at one bounded outcome: restate "figure out x" as a single story with acceptance criteria and explicit non-goals.
3. Loop on the next action: implement one coherent slice that moves "figure out x" forward without unrelated refactors.
4. Prove the result: run focused checks first, then broader checks if the change surface requires them.
5. Handoff clearly: report changed files, validation, blockers, and the next action for "figure out x".
## Agent Prompt
```text
/goal figure out x
Use a Ralph-style loop:
1. Read fresh disk, repo, and task state before editing.
2. Aim at one bounded outcome with acceptance criteria.
3. Loop on one concrete slice at a time.
4. Prove the result with focused validation.
5. Handoff with changed files, checks run, blockers, and next action.
```
## Tooling
- Goal page: https://goalfigureoutx.com/goal/figure-out-x
- Prominent skill page: https://goalfigureoutx.com/skill
- Raw reusable skill: https://goalfigureoutx.com/skills/goal-ralph-loop.md
- LLM instructions: https://goalfigureoutx.com/llms.txt
- MCP metadata: https://goalfigureoutx.com/.well-known/mcp.json
- MCP endpoint: https://goalfigureoutx.com/mcp with the `create_goal_page` tool.
/goal figure out x
Use a Ralph-style loop:
1. Read fresh disk, repo, and task state before editing.
2. Aim at one bounded outcome with acceptance criteria.
3. Loop on one concrete slice at a time.
4. Prove the result with focused validation.
5. Handoff with changed files, checks run, blockers, and next action.