/cf-research

Context footprint: ⚡⚡ (medium) — what does this mean?

The /cf-research skill performs thorough technical research using web search and multiple subagents working in parallel. Results are saved to docs/research/ for future reference.

Usage

/cf-research [topic]

What It Does

1. Scopes the Research — Parses your topic, determines research type, and defines 3-5 key questions
2. Plans Structure — Breaks the topic into logical parts that can be researched independently
3. Parallel Search — Launches multiple subagents to research each part simultaneously
4. Synthesizes Findings — Creates a summary document linking all parts with key findings
5. Saves to Research Dir — Stores in docs/research/<topic-slug>/ for future use

Examples

/cf-research best practices for JWT refresh tokens

Research Types

The skill adapts its approach based on what you're researching:

• Web topic — A technology, library, pattern, or concept (uses WebSearch + WebFetch)
• Codebase — A git repo, folder, or project (uses Read, Glob, Grep)
• Comparison — Comparing multiple options or approaches (each option gets its own doc)

Output Structure

Research output uses numbered part files with a summary:

docs/research/<topic-slug>/
  ├── _summary.md          # Overall summary with links to all parts
  ├── 01-<part-name>.md    # Part 1 findings
  ├── 02-<part-name>.md    # Part 2 findings
  ├── 03-<part-name>.md    # Part 3 findings
  └── ...

Each part document is self-contained and readable independently. The _summary.md provides a high-level overview, links to each part, highlights key findings, and lists open questions.

Parallel Subagents

Research leverages multiple subagents that work simultaneously. Each subagent:

• Focuses on one specific part of the topic
• Uses WebSearch and WebFetch for web topics, or file tools for codebases
• Writes its findings directly to a numbered part file
• Includes code examples, source links, and specific details

When to Use

• Choosing new technologies or frameworks
• Deep-diving into complex topics before implementation
• Security-critical decisions (auth, encryption, data handling)
• Performance optimization research
• Evaluating library alternatives
• Understanding emerging best practices

Output Quality

Each part document follows a consistent template:

• Key Questions — What this part answers
• Findings — Detailed analysis organized by subtopic
• Code Examples — Relevant snippets when applicable
• Sources — Full citations with URLs
• Notes — Additional observations and caveats

The summary document includes:

• High-level overview of the entire research
• Key findings (top 3-5)
• Links to all part documents
• Open questions for further investigation
• Recommended next steps (e.g., "run /cf-plan to plan implementation")