ralph-loop-help

An agent skill by cursor, from cursor/plugins. Tags: documentation, education, explanation, workflow.

What it does

Explain the Ralph Loop plugin, how it works, and available skills. Use when the user asks for help with ralph loop, wants to understand the technique, or needs usage examples.

Install

With the skills CLI, which installs into Claude Code, Codex, Cursor and other agents:

npx skills add cursor/plugins --skill ralph-loop-help

Or copy the skill folder into Claude Code's skills directory by hand (~/.claude/skills for every project, or .claude/skills inside one):

git clone --depth 1 https://github.com/cursor/plugins
cp -r plugins/ralph-loop/skills/ralph-loop-help ~/.claude/skills/ralph-loop-help

Safety box score

Grade B, low risk: 16 of 100 risk points, rated 2026-10-20. The grade rates what the skill and its scripts can reach on the machine of whoever installs it, across eight categories from shell execution to secrets access. How the score works.

This is a documentation-only skill for explaining Ralph Loop, not a script or installer. The main risk is procedural: it describes a looping workflow that can keep re-running an agent until a promise tag or iteration cap is reached, and it uses a workspace state file under `.cursor/ralph/`.

Category levels, from 0 (none) to 3 (high)
Category	Level	Why
Code & shell execution	0 (none)	No shell commands, scripts, or code execution are shipped in `SKILL.md`; it is explanatory text only.
Network access	0 (none)	The file only names external reference URLs (`https://ghuntley.com/ralph/` and a GitHub repo); it does not download or transmit data itself.
Filesystem reach	1 (low)	It describes creating and removing `.cursor/ralph/scratchpad.md`, which writes inside the working tree but stays scoped to the project.
Secrets & credentials	0 (none)	There is no mention of tokens, passwords, keychains, `.env` files, or other secret stores.
Privilege & persistence	0 (none)	Nothing here requires sudo, global installs, services, scheduled tasks, or OS/shell configuration changes.
Instruction hijack surface	1 (low)	It contains user-facing prompt text like `Start a ralph loop:` and `<promise>...`, but no hidden or instruction-sabotaging content.
Irreversible actions	2 (moderate)	The described loop re-feeds the same prompt until completion and says that without a promise tag or `--max-iterations`, `Ralph runs indefinitely`.
Transparency	0 (none)	The skill is plain markdown with readable text; no binaries, installers, or decoded runtime content are present.

Before installing

Confirm you are comfortable with an agent loop that may repeat until a promise tag or max-iteration cap is hit.
Check whether `.cursor/ralph/scratchpad.md` is acceptable to add/remove in your workspace.
Open the linked Ralph docs yourself if you want to verify the external technique description before using it.

Rated 2026-10-20 by an automated review of 1 file from the skill's repository.

Source

Repository: cursor/plugins (all skills from this repository)
Path: ralph-loop/skills/ralph-loop-help/SKILL.md
Branch: main
Collection: cursor
Updated: 2026-09-23

teach — Teach the user a new skill or concept, within this workspace.
teach — Explain a body of work plainly so a person actually understands it. Runs the `how` and `why` skills and weaves what they find into one clear explanation.
scaffold-exercises — Create exercise directory structures with sections, problems, solutions, and explainers that pass linting.
automate-me — Use for "automate me", "create/update/refresh my -mode skill", "turn/capture my preferences or working style into a skill".
continual-learning — Orchestrate continual learning by delegating transcript mining and AGENTS.md updates to `agents-memory-updater`.
recall — Reconstruct your recent working context from your own chat history, live state, and the shared record (user reports, prior fixes, incidents).