Skip to content
Agentmart
Submit a tool

ralph

Autonomous execution loop that never stops until task completion with architect verification

Skill by MaxFerAlten

IndexedMIT
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/maxferalten-ralph

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Autonomous execution loop that never stops until task completion with architect verification

Use cases: productivity

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/ralph

What's inside

Skill name
ralph
Description
Autonomous execution loop that never stops until task completion with architect verification
License
MIT
Instructions
47 lines in SKILL.md · view on GitHub

Files (2)

  • RALPH.md
  • SKILL.md

SKILL.md (first lines)

# Ralph - Autonomous Execution Loop

Ralph is a persistence loop that keeps working on a task until it is fully complete and architect-verified.

## When to Use
- User says "ralph", "don't stop", "keep going", "must complete"
- Task requires guaranteed completion with verification
- Work may span multiple iterations

## Pipeline

### Phase 1: Context Intake
1. Create context snapshot at `.tenderclaw/state/ralph/{task-slug}-{timestamp}.md`
2. Include: task statement, desired outcome, known facts, constraints

### Phase 2: Execution Loop
1. **Delegate in parallel** to specialist agents
2. **Run long operations in background** (builds, tests)
3. **Verify completion** with fresh evidence:
   - Run tests and read output
   - Run build and confirm success
   - Check for 0 errors

### Phase 3: Architect Verification
- STANDARD tier: <5 files, <100 lines
- THOROUGH tier: >20 files or security changes

### Phase 4: Cleanup Pass
- Run code cleanup on changed files
- Re-verify tests still pass

### Phase 5: Completion
- Cancel mode and cleanup state files

## State Files
- `.tenderclaw/state/ralph/{task-slug}.md` - Current state
- `.tenderclaw/state/ralph/progress.json` - Progress tracking

## Verification Checklist
- [ ] All requirements met (no scope reduction)
- [ ] Zero pending TODO items
- [ ] Fresh test run: all pass
- [ ] Fresh build: success
- [ ] Architect verification: APPROVED
- [ ] Cleanup pass completed
- [ ] Post-cleanup regression: pass

From MaxFerAlten/tenderclaw, MIT.

Signals

Stars
0
Forks
0
Open issues
0
Last commit (this folder)
Apr 7, 2026 (6 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from MaxFerAlten

Cite this page

ralph skill. Agent Mart. https://agentmart.dev/skills/maxferalten-ralph (last updated 2026-10-24)