ralph-prompt-generatorlisted

# Ralph Prompt Generator Companion skill for `$ralph-wiggum-codex`. This skill does not exist to synthesize flags first. Its primary job is to improve the prompt itself through a staged prompt-improvement workflow, then finish with a small Ralph-specific handoff. ## Core Rule Clarify, do not prescribe. The receiving agent is equally capable. Improve the briefing, preserve the user's boundaries, and avoid doing the downstream agent's job. ## When To Use Use this skill when: - The source prompt is ambiguous, bloated, or structurally weak. - The user wants a reusable prompt for `$ralph-wiggum-codex`. - The task needs critique and revision, not just runner flag suggestions. - The user wants saved planning, draft, and final prompt files in the repo. Do not use this skill when: - The user already has a production-ready Ralph prompt. - The user wants `$ralph-wiggum-codex` executed immediately with no prompt rewrite. - The only missing detail is a minor flag choice that can be handled inline. ## Inputs Primary input: - `<user_prompt>` Optional supporting context: - `<examples>` - `<feedback>` Treat `<examples>` as demonstrations, not new requirements. Treat `<feedback>` as revision guidance, not permission to weaken the original constraints. ## Repo-Backed Workspace Use this workspace for all saved outputs: - `docs/prompt-improver-spec/artifacts/implementation_plan.md` - `docs/prompt-improver-spec/artifacts/task.md` - `docs/prompt-improver-spec/artifacts/walkthrough.md`