Generated banner for Rpi
Generated logo for Rpi

Rpi

HumanLayerOfficialPublished by HumanLayer

  • Skill
  • Workflow
  • Software development
  • MIT
  • No issues found

From the author

Only use when the user explicitly invokes this skill by name.
— the description by HumanLayer

AutoGineer summary

Written by AutoGineer from the skill’s own files, not by the author

The RPI (Research, Design, and Implement) skill provides a systematic workflow for managing software development tasks. It guides users through a series of defined phases—from initial research and design discussions to PRD creation, technical design, and final implementation. By enforcing a consistent directory structure and documentation standard, it ensures that project context, design decisions, and implementation plans are clearly tracked and easily accessible throughout the development lifecycle.

What it does

  • Facilitates structured research to define project requirements and technical constraints.
  • Manages design discussions by presenting options, tradeoffs, and recommendations to the user.
  • Generates standardized documentation for PRDs, TDDs, and implementation outlines.
  • Maintains a task-specific directory for all workflow artifacts and supporting diagrams.
  • Provides a clear path for iterative updates to existing design and requirement documents.
  • Coordinates the transition between development phases with clear handoff instructions.

When to use it

  • When starting a new software feature or complex technical change.
  • When you need to document design decisions and architectural choices clearly.
  • When you want to follow a repeatable, step-by-step process for software development.
  • When you need to track the evolution of a task from research to implementation.

Advantages

  • Ensures consistent documentation and project structure across different tasks.
  • Promotes informed decision-making by presenting clear tradeoffs and recommendations.
  • Keeps design and implementation plans synchronized with the codebase.
  • Reduces ambiguity by requiring explicit user confirmation for design choices.

Usage examples

“rpi”
The agent initiates the RPI workflow, asks for the task description, and prompts the user to choose between research, design, or implementation phases.
“rpi create-design-discussion”
The agent begins the design discussion phase, analyzing existing research and codebase context to draft a design document with proposed architectural choices.

Install

  1. Download the skill as a ZIP:

    https://autogineer.com/agent-skills/humanlayer/rpi/download
  2. Unzip it into the cross-client skills folder of your project:

    mkdir -p .agents/skills && unzip rpi.zip -d .agents/skills

Documentation by HumanLayer

Research, Design, and Implement

Help the user research a change, work through its design, and implement it. Follow their requested step, using the phase instructions below.

$SKILLBASE means the directory containing this SKILL.md.

<rpi-invocation> in templates means the host's skill command: $rpi in Codex, /rpi in Claude Code, or its plugin-qualified name when the host exposes one. Render next commands with that native invocation.

Choose the Step

Follow the user's requested step. Read $SKILLBASE/references/workflow.md when choosing or explaining a flow or what comes next.

For a task description with no requested step, start with research questions. Ask which flow the user wants: research followed by design discussion, PRD, or TDD, or research only. Wait for their choice before starting the phase, then carry it into suggested next commands.

When asking the user to choose a flow, read $SKILLBASE/references/workflow-choice.md and show its box-and-line chart with the question.

Create writes a new document; iterate updates an existing document in place using the user's feedback or resumes the workflow for a partially finished document.

Load the matching process from $SKILLBASE/references/<step>/process.md:

RequestStep
Research questionscreate-research-questions or iterate-research-questions
Researchcreate-research or iterate-research
Design discussioncreate-design-discussion or iterate-design-discussion
Product requirements / PRDcreate-prd or iterate-prd
Technical design / TDDcreate-tdd or iterate-tdd
Implementation outlinecreate-structure-outline or iterate-structure-outline
Implementimplement-outline
Implementation feedback or fixesiterate-implementation

Update an existing document by default; create one when needed or requested. When several match, use the newest and name the chosen file.

Use the Task Directory

Use the directory supplied by the user.

Otherwise, look for existing task directories in .agents/artifacts/ whose names and workflow filenames match the request or current task branch. Reuse an obvious match without asking. If a directory looks close but could belong to a different task, or several directories look plausible, ask the user which to use before writing. Read only documents allowed by the selected phase when checking a match.

When no existing directory fits, use .agents/artifacts/<branch-name> with / replaced by - if the branch identifies the task. On main, or when no branch identifies the task, choose a short kebab-case slug from the request and use .agents/artifacts/<task-slug> without asking for approval. State the selected path and reuse or create the directory as needed.

Save at the selected path. If the sandbox blocks writes there, ask for scoped write access or a writable task directory and wait for the user's choice.

The RPI framework authors recommend not committing the artifacts.

Number new workflow documents using the next available NN- prefix in the task directory. Keep the existing path when updating a document. Save supporting mockups and diagrams beside the documents and link them as ordinary local files.

Workflow document frontmatter includes repos, a list of zero or more repositories with identifier and sha for each entry. Use repos: [] when none apply. Quote string values as needed for valid YAML.

Review the Document

After writing or updating a workflow document, read $SKILLBASE/references/prose.md and $SKILLBASE/references/show-me.md, then review the document against both. Make it shorter, more visual, and easier for a person to digest: cut words and sections that do not earn their place, replace dense prose with the smallest view that makes the point, and keep each visual beside the short text it supports.

Suggest the Next Step

Use the phase's handoff instructions and $SKILLBASE/references/workflow.md to choose what comes next. Users run each phase in a separate context window by default. When suggesting a next command, include the task directory and chosen flow. Say they can run it here, after compaction, or in a new session.

Use artifact filenames to locate the current phase when asked what's next, then read its document for unfinished work and settled choices. Recognize prd or product-requirements, tdd or tech-design, and structure-outline or outline.

Use Subagents

Use the bundled agent instructions for codebase-locator, codebase-analyzer, codebase-pattern-finder, web-search-researcher, and outline-implementer-agent as directed by each phase.

<important if="you do not have an Agent() or Task() or other tool for running subagents"> read the subagent instructions from $SKILLBASE/references/subagents/SUBAGENT_NAME.md </important>

<important if="you have a tool for running subagents or delegated sessions, but the named subagents are not available"> read the subagent instructions from $SKILLBASE/references/subagents/SUBAGENT_NAME.md and pass them into the subagent with your main prompt, with <instructions> REFERENCE_CONTENTS </instructions> <message> YOUR_MESSAGE ON WHAT TO LOOK FOR OR BUILD </message> </important>

Files

references/create-tdd/process.md

Create a Technical Design Document

Turn the available product context into a technical design. Discuss System Design first, then Program Design, with a separate user approval for each.

Follow these steps in order, running the interview in the main conversation.

1. Read the Available Context

Read the selected workflow documents and named sources in full. A PRD is useful when present; a request plus research is enough to start. Reference the inputs rather than repeating them and ask about missing behavior when needed.

Use the research agents to establish facts or find code patterns that affect a decision. Wait for the findings and update the research document when present.

2. Read the Templates and Write a Small Starting Document

Read $SKILLBASE/references/create-tdd/references/tdd_template.md and $SKILLBASE/references/show-me.md.

Write NN-tdd-<description>.md using the next available number. Start with frontmatter, title, empty System Design and Program Design sections, and Patterns to Follow. Fill repos with relevant identifiers and SHAs, or use repos: [].

Open with the first system-design decision: a short explanation, options, tradeoffs, and one question for the user.

3. Discuss System Design

Describe how components interact: services, endpoints, schemas, queues, stores, and external systems. Show current behavior and the proposed change within this section.

Use Mermaid for data flow, control flow, and sequences. Show the contracts that matter: endpoint or message shapes, high-level signatures, and schemas in the project's language or ORM.

Present one decision per message, then wait for the user's answer. Work through clarifications until it is settled. Rework the section once the user decides, keeping the prose and diagrams a coherent description of the architecture.

When a diagram needs more room, save a focused diagram-<description>.html file in the task directory and link it beside its explanation. Use the project's visual conventions and real labels.

4. Get System Design Approval

Ask the user to review System Design from start to finish. Incorporate feedback and get their approval before moving to program design.

At each section review, offer to open the document in VS Code, Cursor, or the user's preferred editor or browser.

5. Discuss Program Design

Show the code shape that implements the system. Include concrete code views in the discussion: call trees, component trees with state and package boundaries, file-tree changes, dependency-injection maps, method signatures, or pseudocode.

When comparing options, show each as its own code block. Use focused diffs for changes to existing shapes and complete target shapes for new code. Choose the views that expose the decision; the outline will carry the detailed file list.

Ask one question at a time, wait for the user's decision, and then rework Program Design in place. Use headers that state the main point, such as "The worker receives a clock so retry timing is testable." Place each view beside its explanation.

Document patterns to follow with source paths and short examples. If a technical choice changes product scope or user experience, discuss it with the user and update the PRD and mockups when present.

6. Get Program Design Approval

Ask the user to review Program Design and approve the code shape. Incorporate their fixes.

7. Read the Final-Answer Template and Respond

After both sections are approved, read $SKILLBASE/references/create-tdd/references/tdd_final_answer_resolved.md. Offer an outline or direct implementation.

Requirements and compatibility

Works with: Claude Code, Claude apps, OpenAI Codex, GitHub Copilot, Cursor, Gemini CLI, Kiro, OpenCode, Goose, Amp, Roo Code, JetBrains Junie, Factory Droid, Devin (ex-Windsurf), Mistral Vibe, OpenHands, Letta

“Works with” means the tool reads the open Agent Skills format; a skill written for one tool may use features others lack.

Security

No issues found by our static scan· No issues found

A static scan is not a guarantee. Skills are third-party content: review a skill before you let an agent use it.

Attribution and licence

Published by
HumanLayer
Publisher
HumanLayer (Official)
Source
humanlayer/skills · plugins/rpi/skills/rpi · commit 4d69d20
Licence
MIT License (redistribution allowed)
  • LICENSE (added by AutoGineer from LICENSE)
  • ATTRIBUTION.md (written by AutoGineer)
Synced

AutoGineer is not affiliated with or endorsed by HumanLayer or HumanLayer.

Download ZIP

The ZIP holds the skill’s files, its licence files and an ATTRIBUTION.md. Download rpi.zip