Research uncertainties before writing code

Auto-Research Skill provides Claude Code with an explicit-consent research path when uncertainties arise during task implementation. When Claude encounters questions about design choices, algorithm details, API usage, or best practices, this skill enables web research and optional ChatGPT consultation, presenting all findings to the user and waiting for approval before writing any code.

Use this skill when Claude Code faces uncertainties during implementation that require external research: unfamiliar API usage patterns, algorithm selection decisions, framework best practices, or design pattern choices. It is particularly valuable when you want transparent, consent-based research that shows you exactly what information will be gathered before code is written.

Do NOT use this skill if you need immediate code generation without research pauses, or if your workflow cannot accommodate approval steps between research and implementation.

You provide explicit consent before research is performed and approval of findings before they inform code generation. The skill presents research results and waits for your go-ahead.

You receive research findings from web sources and optionally ChatGPT. The skill never sends conversation context, files, browser state, or credentials to a third party without your explicit approval of the exact, redacted text.

The skill supports web research for gathering information about APIs, algorithms, and best practices. It includes optional ChatGPT consultation as a research source, always requiring explicit user consent before any query is sent.

This skill is for developers using Claude Code who value transparency and control over external research during implementation. It suits teams with security or privacy requirements that mandate explicit approval before any data leaves the local environment, and developers who want to understand the research basis for code suggestions before accepting them.