Ralph

What it does

When your agent uses it

• Tasks that involve Autonomous loops

Example prompts

• “/ralph”
• “ralph loop”
• “ralph 루프”
• “/ralph”

Requirements

• Pre-approved tools (allowed-tools): Read, Grep, Glob, Bash, Write, Edit, Agent, AskUserQuestion

Workflow steps

2 steps, taken from the step headings in SKILL.md.

1. 1DoD Collection
2. 2Work Execution

What it can do on your machine

Read from SKILL.md and the folder at commit 7cff032. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves these tools, so the agent can use them without asking each time:

  • Read
  • Grep
  • Glob
  • Bash
  • Write
  • Edit
  • Agent
  • AskUserQuestion

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  Shell commands in SKILL.md call:

  • jq
  • npm
  • tsc
  • eslint

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  No URLs in SKILL.md. Its commands use npm, which can reach the network depending on how they are called.

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Ralph loads about 1.6k tokens when it runs. Until then it costs about 118 tokens; SKILL.md has 613 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~118

When it runs · the whole SKILL.md, loaded when a task matches

~1.6k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

• NotePre-approves every shell command (allowed-tools: Bash)SKILL.md

  allowed-tools: Read, Grep, Glob, Bash, Write, Edit, Agent, AskUserQuestion

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from team-attention/hoyeon at commit 7cff032, republished under its MIT licence (© team-attention). 613 words, ~1,588 tokens.

name

ralph

description

Iterative task completion loop with Definition of Done verification. Combines the Ralph Wiggum technique (prompt re-injection via Stop hook) with DoD-based independent verification. Claude proposes DoD criteria, user confirms, then Claude works autonomously. Stop hook re-injects the original prompt and blocks exit until all DoD items are independently verified. "/ralph", "ralph loop", "ralph 루프", "반복 작업", "DoD 루프", "완료 검증 루프", "task loop", "keep going until done"

allowed-tools

Read, Grep, Glob, Bash, Write, Edit, Agent, AskUserQuestion

validate_prompt

Must contain Phase 1 (DoD Collection) and Phase 2 (Work Execution). Must use AskUserQuestion for DoD confirmation. Must write state via hoyeon-cli session set…

Based on your request, here's my proposed Definition of Done:

1. [concrete criterion 1]
2. [concrete criterion 2]
3. [concrete criterion 3]
...

Each item will be independently verified before the task is considered complete.

Bash: SESSION_ID="[session ID from hook]" && mkdir -p "$HOME/.hoyeon/$SESSION_ID/files" && cat > "$HOME/.hoyeon/$SESSION_ID/files/ralph-dod.md" << 'DODEOF'
# Definition of Done

- [ ] [criterion 1]
- [ ] [criterion 2]
- [ ] [criterion 3]
...
DODEOF

Bash: SESSION_ID="[session ID from hook]" && PROMPT=$(cat << 'PROMPTEOF'
[The user's ORIGINAL request/prompt — exactly as they typed it, before any processing]
PROMPTEOF
) && hoyeon-cli session set --sid "$SESSION_ID" --json "$(jq -n \
  --arg prompt "$PROMPT" \
  --arg dod_file "$HOME/.hoyeon/$SESSION_ID/files/ralph-dod.md" \
  --arg created_at "$(date -u +%Y-%m-%dT%H:%M:%SZ)" \
  '{ralph: {prompt: $prompt, iteration: 0, max_iterations: 10, dod_file: $dod_file, created_at: $created_at}}')"

## Ralph Loop Initialized

**Task**: [summary of what you'll do]
**DoD**: [N] criteria
**Max iterations**: [max_iterations]

Starting work. The loop will verify each DoD item independently before allowing completion.

© team-attention, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in skills/ralph of team-attention/hoyeon.

Open the folder on GitHubat commit 7cff032