Agent skill

Ralph Plan

by TechDufus in TechDufus/oh-my-claude

Structured PRD generation with interview, research, and approval workflow.

MITAuto-check: notesProduct & Project Management

Install Ralph Plan

skills CLI
$ npx skills add TechDufus/oh-my-claude --skill ralph-plan -a claude-code

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

GitHub CLI
$ gh skill install TechDufus/oh-my-claude ralph-plan --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/TechDufus/oh-my-claude.git skills-src && mkdir -p .claude/skills && cp -r skills-src/plugins/oh-my-claude/skills/ralph-plan .claude/skills/ralph-plan && 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-plan
GitHub stars
174
Token cost
~2.4k tokens
SKILL.md length
825 words
Files
1
Skills in repo
12
Repo updated
First seen
Licence
MIT

At a glance

Structured PRD generation with interview, research, and approval workflow.

  • Works in 2 steps: Planning → Approval Gate
  • : /ralph-plan <topic
  • SKILL.md covers What is Ralph Plan?, When This Skill Activates, Workflow and Status Checking, plus 5 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Ralph Plan is an agent skill from TechDufus/oh-my-claude. Structured PRD generation with interview, research, and approval workflow. Triggers on: '/ralph-plan <topic', 'create prd', 'generate prd', 'plan this'. Creates comprehensive Product Requirements Document via interview and research.

Its SKILL.md is about 2.4k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Product & Project Management, covering PRD writing. The repository describes itself as: Add ultrawork to any prompt for maximum parallel execution. The licence is MIT.

When your agent uses it

  • : /ralph-plan <topic
  • Tasks that involve PRD writing

Example prompts

  • “/ralph-plan <topic”
  • “create prd”
  • “generate prd”
  • “/ralph-plan”

Requirements

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

Workflow steps

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

  1. Planning
  2. Approval Gate

What it can do on your machine

Read from SKILL.md and the folder at commit ede75db. 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
    • Write
    • Edit
    • Glob
    • Grep
    • Bash
    • Agent
    • TodoWrite
    • AskUserQuestion

    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 json).

    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 Plan loads about 2.4k tokens when it runs. Until then it costs about 61 tokens; SKILL.md has 825 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~61
When it runs · the whole SKILL.md, loaded when a task matches
~2.4k

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: notes

The automated check noted patterns worth knowing about, such as sudo or a known installer.

  • NotePre-approves every shell command (allowed-tools: Bash)SKILL.md
    allowed-tools: Read, Write, Edit, Glob, Grep, Bash, Agent, TodoWrite, 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 TechDufus/oh-my-claude at commit ede75db, republished under its MIT licence (© TechDufus). 825 words, ~2,435 tokens.

Download SKILL.mdSave it as .claude/skills/ralph-plan/SKILL.md (or your agent's skills folder).
name
ralph-plan
description
Structured PRD generation with interview, research, and approval workflow. Triggers on: '/ralph-plan <topic>', 'create prd', 'generate prd', 'plan this'. Creates comprehensive Product Requirements Document via interview and research.
allowed-tools
Read, Write, Edit, Glob, Grep, Bash, Agent, TodoWrite, AskUserQuestion

Ralph Plan Skill

Structured PRD generation. Interview, research, plan, approve.

What is Ralph Plan?

Ralph Plan creates comprehensive Product Requirements Documents through a deliberate process:

  1. Interview Phase - Understand requirements through focused questions
  2. Research Phase - Analyze codebase via Explore/librarian
  3. Planning Phase - Generate structured plan with clear steps
  4. Approval Gate - User reviews, modifies, and approves the final PRD

The result is a battle-tested plan ready for implementation by any method the user chooses.

When This Skill Activates

CategoryTrigger Phrases
Start planning/ralph-plan <topic>, create prd, generate prd, plan this
Check statusralph-plan status
Resumeresume ralph-plan

Workflow

Phase 1: Planning

Create a structured plan through interview and research.

Step 1: Initialize Draft

Create draft at .claude/plans/drafts/{topic-slug}.md with template:

markdown
# Planning Draft: {topic}

## Status
Phase: Interview
Started: {timestamp}

## Requirements
- [to be captured from interview]

## Decisions
- [choices made during interview]

## Research Findings
- [results from Explore/librarian]

## Open Questions
- [unanswered items]
Step 2: Interview

Conduct focused interview (3-5 questions based on complexity):

Core Questions:

  1. "What problem does this solve?" (understand context)
  2. "What's the scope - minimal viable vs complete?" (set boundaries)
  3. "Any constraints, non-goals, or things to avoid?" (define exclusions)

Follow-up Questions (as needed):

  • Technical approach preferences
  • Dependencies on other work
  • Success criteria

Update draft after each answer.

Step 3: Research

Research the codebase before finalizing:

Agent: Explore
Task: Find relevant files for {topic}
Expected: List of files that will be affected
Agent: oh-my-claude:librarian
Task: Read and summarize key files
Expected: Summary of current implementation patterns

Add findings to draft under "## Research Findings".

Step 4: Generate Plan

Transform draft into structured plan with these sections:

markdown
# Plan: {topic}

## Context
[Why this plan exists, what problem it solves]

## Objectives

### Must Have
- [required outcomes]

### Must NOT
- [explicit exclusions and constraints]

## Implementation Steps
1. [Step with specific file references]
2. [Step with specific file references]
3. ...

## Files to Modify
| File | Changes |
|------|---------|
| `path/to/file.ts` | [what changes] |

## Acceptance Criteria
- [ ] [Testable criterion]
- [ ] [Testable criterion]
- [ ] [Testable criterion]

Phase 2: Approval Gate

User must review and approve the plan before finalization.

Present the Plan
  1. Display the complete plan summary to the user
  2. Use AskUserQuestion tool for approval:
json
{
  "questions": [{
    "question": "How would you like to proceed with this PRD?",
    "header": "PRD Status",
    "options": [
      {"label": "Approve", "description": "Finalize and save the PRD"},
      {"label": "Request Changes", "description": "Modify specific sections"},
      {"label": "Cancel", "description": "Abort and keep draft"}
    ],
    "multiSelect": false
  }]
}

CRITICAL: You MUST use the AskUserQuestion tool for approval. Do NOT output text prompts asking the user to "reply with" something.

Handle User Response
User ResponseAction
"Approve"Finalize PRD (move to final location)
"Request Changes"Ask what to change, update plan, re-present with AskUserQuestion
"Cancel"Keep draft, do not finalize
"Other" (custom text)Interpret and act accordingly, then re-prompt if needed

Do NOT interpret ambiguous responses as approval. If unclear, re-prompt with AskUserQuestion.

On Approval
  1. Move plan from draft to final location:

    .claude/plans/drafts/{topic-slug}.md  -->  .claude/plans/{topic-slug}.md
  2. Delete the draft file

  3. Report completion:

    PRD finalized at .claude/plans/{topic-slug}.md
    
    Next steps:
    - Review the plan at your convenience
    - Implement manually, or use your preferred automation
    - Reference the acceptance criteria to track progress

Status Checking

When user says ralph-plan status:

Ralph Plan Status
=================

**Plan:** {topic}
**Phase:** Interview | Research | Draft Complete | Awaiting Approval | Finalized

--- Plan Location ---
Draft: .claude/plans/drafts/{topic-slug}.md
Final: .claude/plans/{topic-slug}.md (if finalized)

--- Plan Contents ---
Steps: {total}
Acceptance Criteria: {total}
Files Affected: {count}

Examples

Example 1: Feature Implementation PRD
/ralph-plan implement user authentication with JWT

Interview Phase:

  • Q: "What problem does this solve?"
  • A: "Users need to log in securely"
  • Q: "What's the scope?"
  • A: "Basic login/logout with token refresh"
  • Q: "Any constraints?"
  • A: "Must use existing user model, no OAuth"

Research Phase:

  • Explore finds: src/models/User.ts, src/routes/index.ts
  • Librarian summarizes: Uses Express, no existing auth middleware

Plan Generated:

  • 6 implementation steps
  • 4 files to modify
  • 4 acceptance criteria

Approval:

Plan presented to user with AskUserQuestion tool. User selects: "Approve"

Output:

PRD finalized at .claude/plans/user-authentication-jwt.md

Example 2: Refactoring PRD with Modifications
/ralph-plan refactor the API to use dependency injection

Interview Phase:

  • Q: "Which services should use DI?"
  • A: "All services in src/services/"
  • Q: "Testing implications?"
  • A: "Should make mocking easier"

Research Phase:

  • Explore finds: 8 service files, current instantiation patterns
  • Librarian summarizes: Direct instantiation, no DI container

Plan Generated:

  • 12 implementation steps
  • Container setup, service refactoring

Approval: User selects "Request Changes" via AskUserQuestion, specifies "change step 5 to use factory pattern"

Updated Plan: Step 5 modified, re-presented with AskUserQuestion

Second Approval: User selects "Approve"

Output:

PRD finalized at .claude/plans/api-dependency-injection.md

Show full SKILL.md (326 more words)Show less
Example 3: Test Coverage PRD
create prd for adding test coverage to src/services/

Interview Phase:

  • Q: "Unit tests only, or integration too?"
  • A: "Both"
  • Q: "Coverage target?"
  • A: "80% minimum"

Research Phase:

  • Explore finds: 6 service files, existing test patterns
  • Librarian summarizes: Jest setup exists, 2 services have tests

Plan Generated:

  • Test file for each untested service
  • Integration test suite
  • Coverage configuration

Approval: User approves

Output:

PRD finalized at .claude/plans/service-test-coverage.md

Decision Matrices

Complexity Detection
SignalComplexityInterview Depth
"just", "simple", "quick"Simple2-3 questions
Specific file mentionedStandard3-4 questions
"redesign", "overhaul"Complex5-6 questions
Multiple systems involvedComplex6+ questions
When to Skip Research
ScenarioSkip?
User provides full context and file pathsYes
Simple config changeYes
New feature in unfamiliar areaNo
Refactoring existing codeNo

Error Handling

User Cancels During Planning
  1. Confirm: "Cancel planning for '{topic}'?"
  2. If confirmed: Keep draft at .claude/plans/drafts/{topic-slug}.md
  3. Report: "Planning cancelled. Draft saved for later."
User Cancels at Approval Gate
  1. Keep plan at draft location
  2. Report: "PRD not finalized. Draft preserved at .claude/plans/drafts/{topic-slug}.md"

Behavior Rules

MUST DO
  • Complete full planning phase before approval gate
  • Display complete plan to user before finalization
  • Use AskUserQuestion tool for approval (NEVER use text prompts)
  • Wait for explicit approval via tool response
  • Move plan from drafts/ to final location on approval
  • Delete draft after successful finalization
MUST NOT
  • Finalize PRD without explicit user approval
  • Use text prompts for approval (ALWAYS use AskUserQuestion tool)
  • Interpret silence or ambiguous responses as approval
  • Skip the interview or research phases
  • Finalize with incomplete or vague plans
  • Delete draft files without confirmation
  • Proceed to any implementation/execution
SHOULD DO
  • Keep interview concise (3-5 questions)
  • Research codebase before finalizing plan
  • Break complex plans into clear steps
  • Include testable acceptance criteria
  • Preserve drafts for resumability

Integration Notes

Relationship to /plan

Ralph Plan uses an enhanced version of the /plan methodology:

  • Same draft structure
  • Same interview approach
  • Same research phase
  • Same final plan format

The difference: Ralph Plan has a formal approval gate and structured workflow.

State Files Location
.claude/
  plans/
    drafts/
      {topic-slug}.md      # During planning (before approval)
    {topic-slug}.md        # After approval (final PRD)

© TechDufus, 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 plugins/oh-my-claude/skills/ralph-plan of TechDufus/oh-my-claude.

Open the folder on GitHubat commit ede75db

Compare with similar skills

Ralph Plan 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 Plan compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Ralph Plan this skillTechDufus/oh-my-claude174—~2.4kAutomated safety check: NotesMIT
CCPM Project Managementautomazeio/ccpm8.4k—~1.1kAutomated safety check: PassMIT
Ralph Tui Create Beadssubsy/ralph-tui2.5k1 repos~2.6kAutomated safety check: PassMIT
Trellis Brainstormanjiemo/SunnyBeach1787 repos~4kAutomated safety check: PassApache-2.0
Adversarial Speczscole/adversarial-spec5561 repos~8.3kAutomated safety check: NotesMIT
Ralph Tui Create Beads Rustsubsy/ralph-tui2.5k1 repos~2.8kAutomated safety check: PassMIT

Similar skills

  • Runs a spec-driven workflow from PRD to epic to GitHub issues to parallel agents, with status, standup and blocked-work reports from bundled scripts.

    8.4k GitHub stars~1.1k tokensUpdated 6 mo ago
    Product & Project ManagementAuto-check passed
  • Ralph Tui Create Beads

    subsy/ralph-tui

    Convert PRDs to beads for ralph-tui execution. An agent skill from subsy/ralph-tui.

    2.5k GitHub starsUsed in 1 repo~2.6k tokens
    Product & Project ManagementAuto-check passed
  • Trellis Brainstorm

    anjiemo/SunnyBeach

    Guides collaborative requirements discovery before implementation.

    178 GitHub starsUsed in 7 repos~4k tokens
    Product & Project ManagementAuto-check passed
  • Adversarial Spec

    zscole/adversarial-spec

    Iteratively refine a product spec by debating with multiple LLMs (GPT, Gemini, Grok, etc.) until all models agree.

    556 GitHub starsUsed in 1 repo~8.3k tokens
    Product & Project ManagementAuto-check: notes
  • Convert PRDs to beads for ralph-tui execution using beads-rust (br CLI).

    2.5k GitHub starsUsed in 1 repo~2.8k tokens
    Product & Project ManagementAuto-check passed
  • To Issues

    ywwynm/EverythingDone

    Break a plan, spec, or PRD into independently-grabbable issues on the project issue tracker using tracer-bullet vertical slices.

    144 GitHub starsUsed in 12 repos~893 tokens
    Product & Project ManagementAuto-check passed

More from TechDufus/oh-my-claude

All 12 skills in this repo
  • Debugger

    TechDufus/oh-my-claude

    Systematic debugging methodology for diagnosing failures and root cause analysis.

    174 GitHub stars~1k tokensUpdated 2 mo ago
    Auto-check passed
  • Oh My Claude Runtime

    TechDufus/oh-my-claude

    A skill your agent uses when applying oh-my-claude workflow guidance, ultrawork modes, agent orchestration, plan execution, or plugin configuration.

    174 GitHub stars~4.3k tokensUpdated 2 mo ago
    Auto-check passed
  • PR Creation

    TechDufus/oh-my-claude

    MUST be used when creating pull requests. An agent skill from TechDufus/oh-my-claude.

    174 GitHub stars~1.8k tokensUpdated 2 mo ago
    Auto-check passed
  • Ralph Loop Init

    TechDufus/oh-my-claude

    Transform approved plans into ralph loop infrastructure. An agent skill from TechDufus/oh-my-claude.

    174 GitHub stars~4.6k tokensUpdated 2 mo ago
    Auto-check: notes
  • Writing Skills

    TechDufus/oh-my-claude

    Methodology for creating effective skills using TDD principles.

    174 GitHub stars~1.3k tokensUpdated 2 mo ago
    Auto-check passed
  • Init Deep

    TechDufus/oh-my-claude

    Initialize or migrate to nested CLAUDE.md structure for progressive disclosure.

    174 GitHub stars~5.3k tokensUpdated 2 mo ago
    Auto-check: notes

Questions about Ralph Plan

What does Ralph Plan do?

Structured PRD generation with interview, research, and approval workflow. Ralph Plan is an agent skill from TechDufus/oh-my-claude. Structured PRD generation with interview, research, and approval workflow.

When should I use Ralph Plan?

Ralph Plan fits situations like: : /ralph-plan <topic; tasks that involve PRD writing.

How do I install Ralph Plan in Claude Code?

Run `npx skills add TechDufus/oh-my-claude --skill ralph-plan -a claude-code`. Or copy the skill folder (plugins/oh-my-claude/skills/ralph-plan in TechDufus/oh-my-claude) into .claude/skills/ralph-plan in your project. Claude Code loads it when a task matches its description.

How do I install Ralph Plan in Codex?

Run `npx skills add TechDufus/oh-my-claude --skill ralph-plan -a codex`. Or copy the skill folder (plugins/oh-my-claude/skills/ralph-plan in TechDufus/oh-my-claude) into .agents/skills/ralph-plan in your project. Codex loads it when a task matches its description.

Can I use Ralph Plan 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 TechDufus/oh-my-claude --skill ralph-plan -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-plan, .gemini/skills/ralph-plan, .github/skills/ralph-plan and .opencode/skills/ralph-plan in your project.

What does Ralph Plan need to run?

SKILL.md names no scripts, command-line tools or credentials: Ralph Plan is instructions for the agent only. Its frontmatter pre-approves these tools: Read, Write, Edit, Glob, Grep, Bash, Agent, TodoWrite, AskUserQuestion.

Does Ralph Plan 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 Plan safe to install?

Our automated static check of SKILL.md found notes only (pre-approves every shell command (allowed-tools: bash)), nothing it rates as a warning. It is not a guarantee. Review the folder before installing.

What licence does Ralph Plan use?

Ralph Plan 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 Plan use?

About 2.4k tokens (SKILL.md is roughly 9.7k 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 Plan?

Skills that share tags, products or a category with Ralph Plan: CCPM Project Management (automazeio/ccpm, 8.4k stars), Ralph Tui Create Beads (subsy/ralph-tui, 2.5k stars), Trellis Brainstorm (anjiemo/SunnyBeach, 178 stars) and Adversarial Spec (zscole/adversarial-spec, 556 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Ralph Plan?

TechDufus (a GitHub user) maintains it in TechDufus/oh-my-claude, which has 174 GitHub stars. The repository holds 12 skills in this directory. The repository was last updated on July 13, 2026.

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