From paxdev
Research a codebase by spawning parallel sub-agents and synthesizing findings
npx claudepluginhub paxsonsa/myskillz --plugin paxdev# 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. ## Initial Setup: The research request for the user MAY be provided at the end of this instruction inside of `<prompt></prompt>` When this command is invoked: If the prompt tag is empty respond with: If the prompt tag is populated, proceed with using it as the research request. Then wait for the user's research query. ## Steps to follow after receiving the research query: 1. **Read any directly mentio...
/researchPerforms adaptive deep web research on a query with configurable --depth and --strategy options. Outputs markdown report with executive summary, analysis, confidence scores, and cited sources.
/researchPrompts for research intensity (quick/standard/deep) then invokes octo:discover skill for multi-AI orchestration, synthesis, and analysis.
/researchConducts multi-turn deep research on a codebase topic over 5 iterations, tracing code paths with citations, Mermaid diagrams, tables, and confidence ratings.
/researchConducts institutional-grade equity research analysis on a stock ticker via web searches, producing a structured report with summary, financials, catalysts, valuation, risks, and technicals.
/researchGathers knowledge from trusted web sources and files, cross-references claims across multiple sources, and produces cited research documents in docs/research/. Optional skill distillation via --skill-for.
/researchRuns research phase for current or specified spec: gathers context, optional interview, parallel subagent research, synthesizes research.md, reviews, approves, and finalizes state.
You are tasked with conducting comprehensive research across the codebase to answer user questions by spawning parallel sub-agents and synthesizing their findings.
The research request for the user MAY be provided at the end of this instruction inside of <prompt></prompt>
When this command is invoked:
If the prompt tag is empty 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.
If the prompt tag is populated, proceed with using it as the research request.
Then wait for the user's research query.
Read any directly mentioned files first:
Check for existing Discovery Context (Preflight): Before deep-diving into code, check if high-level repo context exists.
discovery-locator to find docs in .llm/shared/context/.discovery-analyzer on the best match with the user's query.proceed_using_discovery: Read the discovery doc and use it as baseline understanding of architecture/components.proceed_but_verify_live_code: Read the doc but verify critical details against live code.request_discovery_refresh: Warn the user that discovery is stale/missing. Ask if they want to run /discovery first, or proceed with caution.Why this matters: Discovery docs answer "What is this system?" and "Where does X live conceptually?" — letting you focus research on "How does it actually work in code right now?"
Analyze and decompose the research question:
Spawn parallel sub-agent tasks for comprehensive research:
The key is to use these agents intelligently:
Wait for all sub-agents to complete and synthesize findings:
Gather metadata for the research document:
.llm/shared/research/YYYY-MM-DD-ENG-XXXX-description.md
YYYY-MM-DD-ENG-XXXX-description.md where:
2025-01-08-ENG-1478-parent-child-tracking.md2025-01-08-authentication-flow.mdGenerate research document:
---
date: [Current date and time with timezone in ISO format]
researcher: [Researcher name]
git_commit: [Current commit hash]
branch: [Current branch name]
repository: [Repository name]
topic: "[User's Question/Topic]"
tags: [research, codebase, relevant-component-names]
status: complete
last_updated: [Current date in YYYY-MM-DD format]
last_updated_by: [Researcher name]
---
# Research: [User's Question/Topic]
**Date**: [Current date and time with timezone from step 4]
**Researcher**: [Researcher name]
**Git Commit**: [Current commit hash from step 4]
**Branch**: [Current branch name from step 4]
**Repository**: [Repository name]
## Research Question
[Original user query]
## Summary
[High-level findings answering the user's question]
## Detailed Findings
### [Component/Area 1]
- Finding with reference ([file.ext:line](link))
- Connection to other components
- Implementation details
### [Component/Area 2]
...
## Code References
- `path/to/file.py:123` - Description of what's there
- `another/file.ts:45-67` - Description of the code block
## Architecture Insights
[Patterns, conventions, and design decisions discovered]
## Historical Context (from .llm/)
[Relevant insights from .llm/ directory with references]
- `.llm/shared/something.md` - Historical decision about X
- `.llm/local/notes.md` - Past exploration of Y
Note: Paths exclude "searchable/" even if found there
## Related Research
[Links to other research documents in .llm/shared/research/]
## Open Questions
[Any areas that need further investigation]
Add GitHub permalinks (if applicable):
git branch --show-current and git statusgh repo view --json owner,namehttps://github.com/{owner}/{repo}/blob/{commit}/{file}#L{line}Sync and present findings:
Handle follow-up questions:
last_updated and last_updated_by to reflect the updatelast_updated_note: "Added follow-up research for [brief description]" to frontmatter## Follow-up Research [timestamp].llm/searchable/allison/old_stuff/notes.md → thoughts/allison/old_stuff/notes.md.llm/searchable/shared/prs/123.md → thoughts/shared/prs/123.md.llm/searchable/global/shared/templates.md → thoughts/global/shared/templates.mdlast_updated, git_commit)