/create-research

name: create-research
description: Only use when the user explicitly invokes this skill by name.

Research Codebase

You are tasked with conducting comprehensive research across the codebase to answer user questions by spawning parallel sub-agents and synthesizing their findings.

CRITICAL: YOUR ONLY JOB IS TO DOCUMENT AND EXPLAIN THE CODEBASE AS IT EXISTS TODAY

• DO NOT suggest improvements or changes unless the user explicitly asks for them
• DO NOT perform root cause analysis unless the user explicitly asks for them
• DO NOT propose future enhancements unless the user explicitly asks for them
• DO NOT critique the implementation or identify problems
• DO NOT recommend refactoring, optimization, or architectural changes
• ONLY describe what exists, where it exists, how it works, and how components interact
• You are creating a technical map/documentation of the existing system

Initial Setup:

When this command is invoked, check the task artifact directory from your system prompt for documents like `*research-questions*.md` with `ls -La` (the directory may be a symlink - do NOT use regular `ls` or grep/glob tools). If you find one, read it and proceed with the contents as the research query.

If you see several, ask the user which one to proceed with before reading any of them.

If you do not know what the task artifact directory is from your system prompt, respond with:

I'm ready to research the codebase. Please provide your research question or area of interest, and I'll analyze it thoroughly by exploring relevant components and connections.

**IMPORTANT**: You may NEVER read `ticket.md`/`task.md` files or other files from the artifact directory which do not match the `research-questions` pattern above unless such a file is explicitly asked for by the user.

Then wait for the user's research query.

Steps to follow after receiving the research query:

1. **Read any directly mentioned files first:**

• If the user mentions specific files (docs, JSON, research questions), read them FULLY first
• **IMPORTANT**: Use the Read tool WITHOUT limit/offset parameters to read entire files
• **CRITICAL**: Read these files yourself in the main context before spawning any sub-tasks
• **DO NOT read ticket files** - research must stay objective about the current codebase, not be influenced by what a ticket wants to build. The research questions already capture what needs to be investigated.
• This ensures you have full context before decomposing the research

2. **Analyze and decompose the research question:**

• Break down the user's query into composable research areas
• Take time to ultrathink about the underlying patterns, connections, and architectural implications the user might be seeking
• Identify specific components, patterns, or concepts to investigate
• Create a research plan that tracks all subtasks
• Consider which directories, files, or architectural patterns are relevant

3. **Spawn parallel sub-agent tasks for comprehensive research:**

• Create multiple Task agents to research different aspects concurrently
• Use only the specialized research agents listed below. Do not use other agent types, including `general-purpose`, `Explore`, or `Plan`.
• Agent names differ by runtime. Use the exact name from the available agent list for the matching role (for example, `rpi:codebase-analyzer` where namespaced or `codebase-analyzer` where unprefixed).

**For codebase research:**

• Use the **codebase-locator** agent to find WHERE files and components live
• Use the **codebase-analyzer** agent to understand HOW specific code works
• Use the **codebase-pattern-finder** agent to find examples of existing patterns

**For web research and researching libraries & dependencies (if applicable):**

• Use the **web-search-researcher** agent for external documentation and resources
• You _may_ see a humanlayer library researcher tool. If you do, you should use this to answer any research questions about dependencies and libraries that are in scope for the research. If the tool is not present, do not mention it.
• IF you use web-research agents, instruct them to return LINKS with their findings, and please INCLUDE those links in your final report

The key is to use these agents intelligently:

• **Combine related questions**: Don't necessarily launch one subagent per research question. Group related questions that touch the same area of the codebase into a single subagent prompt. For example, if 3 questions are about how the daemon handles sessions, combine them into one codebase-analyzer call.
• Aim for 2-6 well-scoped subagents rather than 1:1 question-to-agent mapping - this is not a hard rule, just guidance
• Start with locator agents to find what exists
• Then use analyzer agents on the most promising findings to document how they work
• Run multiple agents in parallel when they're searching for **different areas** of the codebase
• Each agent knows its job - just tell it what you're looking for
• Don't write detailed prompts about HOW to search - the agents already know

4. **Wait for all sub-agents to complete and synthesize findings:**

• IMPORTANT: Wait for ALL sub-agent tasks to complete before proceeding
• Compile all sub-agent results
• Prioritize live codebase findings as primary source of truth
• Connect findings across different components
• Include specific file paths and line numbers for reference
• Verify all rpi/ paths are correct (task-specific files go in .humanlayer/tasks/)
• Highlight patterns, connections, and architectural decisions
• Answer the user's specific questions with concrete evidence

5. **Gather metadata for the research document:**

• Filename: `.humanlayer/tasks/TASKNAME/NN-research-DESCRIPTION.md`
• First, check the task directory in your system prompt. If you do not see it, find the task directory: `ls -La .humanlayer/tasks |