Agent skill

Ralph Prompt Single Task

by majiayu000 in majiayu000/claude-skill-registry

Generate Ralph-compatible prompts for single implementation tasks.

MITAuto-check passedDevelopment

Install Ralph Prompt Single Task

skills CLI
$ npx skills add majiayu000/claude-skill-registry --skill ralph-prompt-single-task -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install majiayu000/claude-skill-registry ralph-prompt-single-task --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/majiayu000/claude-skill-registry.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/ai-llm/ralph-prompt-single-task-adaptationio-skrillz-2 .claude/skills/ralph-prompt-single-task && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
ralph-prompt-single-task
GitHub stars
666
Used in
1 other repo
Token cost
~2.7k tokens
SKILL.md length
419 words
Files
2
Skills in repo
1,273
Repo updated
First seen
Licence
MIT

At a glance

Generate Ralph-compatible prompts for single implementation tasks.

  • Works in 4 steps: Analyze the Task → Define Completion Criteria → Structure the Prompt → …
  • Creating prompts for bug fixes
  • SKILL.md covers Overview, Quick Start, Prompt Generation Workflow and Best Practices, plus 2 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Ralph Prompt Single Task is an agent skill from majiayu000/claude-skill-registry. Generate Ralph-compatible prompts for single implementation tasks. Creates prompts with clear completion criteria, automatic verification, and TDD approach. Use when creating prompts for bug fixes, single features, refactoring tasks, or any focused implementation that can be completed in one session.

Its SKILL.md is about 2.7k tokens, which your agent loads only when the skill is triggered. The skill folder holds 1 other file (for example `metadata.json`).

It sits in Development, covering Debugging, Refactoring and Test-driven development. The repository describes itself as: Searchable Claude Code skills catalog with source-linked guides and generated registry artifacts. The licence is MIT.

When your agent uses it

  • Creating prompts for bug fixes
  • Single features
  • Refactoring tasks
  • Any focused implementation that can be completed in one session

Example prompts

  • “/ralph-prompt-single-task”

Workflow steps

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

  1. Analyze the Task
  2. Define Completion Criteria
  3. Structure the Prompt
  4. Add Verification Commands

What it can do on your machine

Read from SKILL.md and the folder at commit 2d14a69. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md (its code samples are markdown and bash).

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

  • Network

    No URLs in SKILL.md.

    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 Prompt Single Task loads about 2.7k tokens when it runs. Until then it costs about 82 tokens; SKILL.md has 419 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~82
When it runs · the whole SKILL.md, loaded when a task matches
~2.7k

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

Auto-check passed

The automated check found no risky patterns in SKILL.md.

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 majiayu000/claude-skill-registry at commit 2d14a69, republished under its MIT licence (© majiayu000). 419 words, ~2,706 tokens.

Download SKILL.mdSave it as .claude/skills/ralph-prompt-single-task/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.
name
ralph-prompt-single-task
description
Generate Ralph-compatible prompts for single implementation tasks. Creates prompts with clear completion criteria, automatic verification, and TDD approach. Use when creating prompts for bug fixes, single features, refactoring tasks, or any focused implementation that can be completed in one session.

Ralph Prompt Generator: Single Task

Overview

Generates well-structured prompts optimized for the Ralph Wiggum technique for single, focused tasks. These prompts include clear completion criteria, automatic verification (tests), self-correction loops, and proper <promise> completion tags.

Best For:

  • Bug fixes with clear success criteria
  • Single feature implementations
  • Refactoring tasks
  • Adding tests for existing code
  • Documentation for specific modules
  • Performance optimizations with measurable targets

Ralph Philosophy: This generator embraces the principle that failures are deterministic and fixable. Each iteration learns from the previous one. Don't fear failures—they're expected and provide data for improvement through prompt tuning.

Quick Start

Input Required:

  1. Task description (what needs to be done)
  2. Success criteria (how to verify completion)
  3. Completion promise text (the phrase that signals done)

Generate prompt with:

Generate a Ralph prompt for: [task description]
Success criteria: [how to verify]
Promise: [completion phrase]

Prompt Generation Workflow

Step 1: Analyze the Task

Identify these elements:

  • Action: What specific work needs to be done?
  • Scope: What files/modules are affected?
  • Verification: How can success be automatically verified?
  • Edge cases: What could go wrong?
Step 2: Define Completion Criteria

Good completion criteria are:

  • Measurable (tests pass, no errors, specific output)
  • Verifiable automatically (not subjective)
  • Binary (clearly done or not done)

Examples:

Task TypeGood CriteriaBad Criteria
Bug fix"Tests pass, bug no longer reproducible""Bug is fixed"
Feature"All CRUD endpoints return 200, tests pass""Feature works"
Refactor"All tests pass, code uses new pattern""Code is cleaner"
Step 3: Structure the Prompt

Use this template:

markdown
# Task: [Clear Task Title]

## Objective
[1-2 sentence clear description of what needs to be accomplished]

## Context
[Brief background - what exists, why this is needed]

## Requirements
1. [Specific requirement 1]
2. [Specific requirement 2]
3. [Specific requirement 3]

## Success Criteria
- [ ] [Measurable criterion 1]
- [ ] [Measurable criterion 2]
- [ ] [Measurable criterion 3]

## Verification Steps
1. [How to verify criterion 1]
2. [How to verify criterion 2]
3. [How to verify criterion 3]

## TDD Approach
1. Write/identify failing tests for the requirement
2. Implement the solution
3. Run tests: `[test command]`
4. If tests fail, analyze output and fix
5. Repeat until all tests pass

## Completion
When ALL of the following are true:
- All success criteria met (checklist above)
- All tests passing
- No regressions introduced

Output: <promise>[COMPLETION_PHRASE]</promise>

## If Stuck
After multiple iterations without progress:
- Document what's blocking progress
- List approaches attempted
- Identify specific error or issue
- Do NOT output false promise - continue iterating or wait for max-iterations
Step 4: Add Verification Commands

Include specific commands to verify success:

markdown
## Verification Commands
```bash
# Run tests
npm test
# or
pytest tests/

# Check for errors
npm run lint
# or
ruff check .

# Verify specific functionality
curl -X GET http://localhost:3000/api/endpoint

## Example Prompts

### Example 1: Bug Fix

**Input:**
- Task: Fix authentication token refresh race condition
- Verification: Auth tests pass, no 401 errors in logs
- Promise: AUTH_FIX_COMPLETE

**Generated Prompt:**

```markdown
# Task: Fix Authentication Token Refresh Race Condition

## Objective
Fix the race condition in token refresh that causes intermittent 401 errors when multiple requests trigger refresh simultaneously.

## Context
The auth module in `src/auth/` handles JWT token refresh. When a token expires, multiple concurrent requests may all attempt to refresh, causing a race condition where some requests fail with 401.

## Requirements
1. Implement token refresh locking/queuing mechanism
2. Ensure concurrent requests wait for single refresh
3. Maintain backward compatibility with existing auth flow
4. Add tests covering concurrent refresh scenario

## Success Criteria
- [ ] All existing auth tests pass
- [ ] New concurrent refresh test passes
- [ ] No 401 errors in integration test logs
- [ ] Token refresh happens only once per expiration

## Verification Steps
1. Run auth tests: `npm test -- --grep "auth"`
2. Run concurrent refresh test: `npm test -- --grep "concurrent"`
3. Check logs for 401 after integration test
4. Verify single refresh in logs during concurrent requests

## TDD Approach
1. Write test for concurrent token refresh scenario
2. Verify test fails with current implementation
3. Implement locking mechanism
4. Run tests: `npm test`
5. If tests fail, analyze and fix
6. Repeat until all pass

## Completion
When ALL of the following are true:
- All auth tests passing (including new concurrent test)
- Integration tests show no 401 errors
- Single refresh verified in logs

Output: <promise>AUTH_FIX_COMPLETE</promise>

## If Stuck
After multiple iterations:
- Document the specific race condition behavior
- List locking approaches tried
- Identify where locks are failing
- Do NOT output false promise
Show full SKILL.md (173 more words)Show less
Example 2: Feature Implementation

Input:

  • Task: Add pagination to /api/users endpoint
  • Verification: Pagination params work, tests pass
  • Promise: PAGINATION_DONE

Generated Prompt:

markdown
# Task: Add Pagination to Users Endpoint

## Objective
Implement offset-based pagination for GET /api/users endpoint with page, limit, and total count.

## Context
The users endpoint in `src/routes/users.ts` currently returns all users. As the user base grows, this needs pagination for performance.

## Requirements
1. Accept `page` (default 1) and `limit` (default 20, max 100) query params
2. Return paginated response with metadata: { data: [], page, limit, total, totalPages }
3. Handle edge cases: invalid params, page beyond data, empty results
4. Add comprehensive tests for pagination

## Success Criteria
- [ ] GET /api/users returns paginated response
- [ ] `page` and `limit` query params work correctly
- [ ] Response includes correct metadata (total, totalPages)
- [ ] Edge cases handled gracefully
- [ ] All new pagination tests pass
- [ ] Existing user tests still pass

## Verification Steps
1. Run tests: `npm test -- --grep "users"`
2. Manual verification:
   - `curl localhost:3000/api/users?page=1&limit=10`
   - `curl localhost:3000/api/users?page=999&limit=10`
   - `curl localhost:3000/api/users?limit=200`

## TDD Approach
1. Write tests for pagination behavior
2. Verify tests fail
3. Implement pagination logic
4. Run tests: `npm test`
5. If failures, fix and retry
6. Test edge cases manually

## Completion
When ALL of the following are true:
- All pagination tests pass
- All existing user tests pass
- Manual verification shows correct behavior
- Edge cases handled (returns empty for page beyond data)

Output: <promise>PAGINATION_DONE</promise>

## If Stuck
Document:
- Which tests are failing and why
- What pagination approach was tried
- Specific error messages
- Do NOT output false promise
Example 3: Refactoring Task

Input:

  • Task: Convert callbacks to async/await in database module
  • Verification: All DB tests pass, no callback patterns remain
  • Promise: ASYNC_REFACTOR_COMPLETE

Generated Prompt:

markdown
# Task: Convert Database Module to Async/Await

## Objective
Refactor all callback-based database operations in `src/db/` to use async/await pattern for improved readability and error handling.

## Context
The database module uses callback patterns from legacy code. Modern async/await improves code clarity and makes error handling more consistent.

## Requirements
1. Convert all callback-based functions to async/await
2. Replace callback error handling with try/catch
3. Update all callers to use await
4. Maintain identical external API behavior
5. No callback patterns remaining in db module

## Success Criteria
- [ ] All database functions use async/await
- [ ] No callback patterns in src/db/*.ts files
- [ ] All existing database tests pass
- [ ] All integration tests pass
- [ ] No TypeScript errors

## Verification Steps
1. Run tests: `npm test -- --grep "database"`
2. Search for callbacks: `grep -r "callback\|cb)" src/db/`
3. TypeScript check: `npm run typecheck`
4. Integration tests: `npm run test:integration`

## TDD Approach
1. Run all tests - ensure baseline passes
2. Convert one function at a time
3. Run tests after each conversion
4. If failures, fix before continuing
5. After all conversions, verify no callbacks remain

## Completion
When ALL of the following are true:
- All database tests passing
- grep for callbacks returns empty
- TypeScript compilation succeeds
- Integration tests pass

Output: <promise>ASYNC_REFACTOR_COMPLETE</promise>

## If Stuck
Document:
- Which function is causing issues
- The specific error or test failure
- What conversion approach was tried
- Do NOT output false promise

Best Practices

DO:
  • Use specific, measurable success criteria
  • Include actual test commands
  • Add verification steps for each criterion
  • Include TDD approach for self-correction
  • Set clear completion conditions
  • Provide "If Stuck" guidance
DON'T:
  • Use vague criteria ("make it work")
  • Skip verification commands
  • Forget edge cases
  • Make subjective success criteria
  • Allow escape without genuine completion

Integration with Ralph Loop

Run your generated prompt:

bash
/ralph-wiggum:ralph-loop "[paste generated prompt]" --completion-promise "YOUR_PROMISE" --max-iterations 30

Recommended settings:

  • Bug fixes: --max-iterations 15-20
  • Features: --max-iterations 25-35
  • Refactoring: --max-iterations 20-30

Validation Checklist

Before using a generated prompt, verify:

  • Task has single, focused objective
  • Success criteria are measurable
  • Verification commands are specific and runnable
  • TDD approach is defined
  • Completion promise is clear and specific
  • "If Stuck" section provides guidance

For multi-task prompts, see ralph-prompt-multi-task. For project-level prompts, see ralph-prompt-project. For research/analysis prompts, see ralph-prompt-research.

© majiayu000, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file in skills/ai-llm/ralph-prompt-single-task-adaptationio-skrillz-2 of majiayu000/claude-skill-registry.

  • SKILL.md
  • metadata.json

Open the folder on GitHubat commit 2d14a69

Used in 1 other repository

We found 2 copies of this SKILL.md (exact, near-identical or edited) in other folders, from 1 other GitHub owner. This page covers the copy in majiayu000/claude-skill-registry, which our catalogue first saw on October 7, 2026.

Compare with similar skills

Ralph Prompt Single Task next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Ralph Prompt Single Task compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Ralph Prompt Single Task this skillmajiayu000/claude-skill-registry6661 repos~2.7kAutomated safety check: PassMIT
Devnpc-live/clawfirm156—~642Automated safety check: PassNone
Pstack Skillfabricioctelles/skills106—~5kAutomated safety check: PassApache-2.0
Development Workflowruby-git/ruby-git1.8k—~6.1kAutomated safety check: PassMIT
Test Driven Developmentsickn33/agentic-awesome-skills47k1 repos~2.1kAutomated safety check: PassMIT
Solidramziddin/solid-skills606—~2.7kAutomated safety check: PassNone

Similar skills

  • Dev

    npc-live/clawfirm

    Software development workflow dispatcher. An agent skill from npc-live/clawfirm.

    156 GitHub stars~642 tokensUpdated 3 mo ago
    DevelopmentAuto-check passed
  • Pstack Skill

    fabricioctelles/skills

    Rigorous engineering orchestrator ported from Lauren Tan's pstack (poteto-mode): reads your task, picks one of 23 playbooks (bug fix, feature, refactoring, perf, investigation, prototype, babysit…

    106 GitHub stars~5k tokensUpdated 4 days ago
    DevelopmentAuto-check passed
  • Development Workflow

    ruby-git/ruby-git

    Follows a strict Test-Driven Development (TDD) workflow with four phases: triage, prepare, execute, and finalize.

    1.8k GitHub stars~6.1k tokensUpdated 7 days ago
    Testing & QAAuto-check passed
  • Test Driven Development

    sickn33/agentic-awesome-skills

    Use a failing behavioral test to guide a feature or bug fix, then implement and refactor with relevant regression checks.

    47k GitHub starsUsed in 1 repo~2.1k tokens
    Testing & QAAuto-check passed
  • Solid

    ramziddin/solid-skills

    A skill your agent uses when writing code, implementing features, refactoring, planning architecture, designing systems, reviewing code, or debugging.

    606 GitHub stars~2.7k tokensUpdated 1 mo ago
    DevelopmentAuto-check passed
  • Code Review Graph Navigator

    handsontable/handsontable

    Queries a pre-built, Tree-sitter-based code graph of the whole monorepo instead of grepping call chains, for exploring, debugging, refactoring or reviewing code.

    22k GitHub stars~939 tokensUpdated yesterday
    DevelopmentAuto-check passed

More from majiayu000/claude-skill-registry

All 1,273 skills in this repo
  • Deep Research

    majiayu000/claude-skill-registry

    Multi-source deep research using firecrawl and exa MCPs. An agent skill from majiayu000/claude-skill-registry.

    666 GitHub starsUsed in 6 repos~1.1k tokens
    Auto-check passed
  • Exa Search

    majiayu000/claude-skill-registry

    Neural search via Exa MCP for web, code, and company research.

    666 GitHub starsUsed in 5 repos~856 tokens
    Auto-check passed
  • Fal AI Media

    majiayu000/claude-skill-registry

    Unified media generation via fal.ai MCP — image, video, and audio.

    666 GitHub starsUsed in 5 repos~1.7k tokens
    Auto-check passed
  • Pyzotero

    majiayu000/claude-skill-registry

    Interact with Zotero reference management libraries using the pyzotero Python client.

    666 GitHub starsUsed in 5 repos~1.6k tokens
    Auto-check: notes
  • Bgpt Paper Search

    majiayu000/claude-skill-registry

    Search scientific papers and retrieve structured experimental data extracted from full-text studies via the BGPT MCP server.

    666 GitHub starsUsed in 4 repos~619 tokens
    Auto-check: notes
  • Bio Alignment Pairwise

    majiayu000/claude-skill-registry

    Perform pairwise sequence alignment using Biopython Bio.Align.PairwiseAligner.

    666 GitHub starsUsed in 4 repos~1.7k tokens
    Auto-check passed

Categories

Questions about Ralph Prompt Single Task

What does Ralph Prompt Single Task do?

Generate Ralph-compatible prompts for single implementation tasks. Ralph Prompt Single Task is an agent skill from majiayu000/claude-skill-registry. Generate Ralph-compatible prompts for single implementation tasks.

When should I use Ralph Prompt Single Task?

Ralph Prompt Single Task fits situations like: creating prompts for bug fixes; single features; refactoring tasks; any focused implementation that can be completed in one session.

How do I install Ralph Prompt Single Task in Claude Code?

Run `npx skills add majiayu000/claude-skill-registry --skill ralph-prompt-single-task -a claude-code`. Or copy the skill folder (skills/ai-llm/ralph-prompt-single-task-adaptationio-skrillz-2 in majiayu000/claude-skill-registry) into .claude/skills/ralph-prompt-single-task in your project. Claude Code loads it when a task matches its description.

How do I install Ralph Prompt Single Task in Codex?

Run `npx skills add majiayu000/claude-skill-registry --skill ralph-prompt-single-task -a codex`. Or copy the skill folder (skills/ai-llm/ralph-prompt-single-task-adaptationio-skrillz-2 in majiayu000/claude-skill-registry) into .agents/skills/ralph-prompt-single-task in your project. Codex loads it when a task matches its description.

Can I use Ralph Prompt Single Task in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add majiayu000/claude-skill-registry --skill ralph-prompt-single-task -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/ralph-prompt-single-task, .gemini/skills/ralph-prompt-single-task, .github/skills/ralph-prompt-single-task and .opencode/skills/ralph-prompt-single-task in your project.

What does Ralph Prompt Single Task need to run?

SKILL.md names no scripts, command-line tools or credentials: Ralph Prompt Single Task is instructions for the agent only.

Does Ralph Prompt Single Task access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Ralph Prompt Single Task safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Ralph Prompt Single Task use?

Ralph Prompt Single Task is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Ralph Prompt Single Task use?

About 2.7k tokens (SKILL.md is roughly 11k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Ralph Prompt Single Task?

Skills that share tags, products or a category with Ralph Prompt Single Task: Dev (npc-live/clawfirm, 156 stars), Pstack Skill (fabricioctelles/skills, 106 stars), Development Workflow (ruby-git/ruby-git, 1.8k stars) and Test Driven Development (sickn33/agentic-awesome-skills, 47k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Ralph Prompt Single Task?

majiayu000 (a GitHub user) maintains it in majiayu000/claude-skill-registry, which has 666 GitHub stars. The repository holds 1,273 skills in this directory. The repository was last updated on October 7, 2026.

Source: majiayu000/claude-skill-registry on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.