Agent skill · Backend & API

codex-search

Search the web using OpenAI Codex CLI to find current documentation, best practices, solutions, and technical information. Use when the user needs to research libraries, find API documentation, troubleshoot errors, or learn about new technologies. Requires Codex CLI installed.

majiayu000github.com/majiayu000GitHub ↗
codexclaude-codecan modify filesMIT
Install
npx skills add majiayu000/claude-skill-registry --skill codex-search --agent codex

Same command for any agent — swap --agent for claude-code, cursor, copilot.

Facts
Files in the skill folder: 2
SKILL.md size: 11 KB
Bundled scripts: none
Allowed tools: BashReadGrepGlobWebFetchWebSearch
Path: skills/analysis/codex-search/SKILL.md
Open the folder on GitHub →
Where it comes from
Stars: 534
Language: HTML

Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.

From the SKILL.md

# Codex Search Skill Use OpenAI Codex CLI with web search capabilities to research current information, documentation, and solutions. This is a **read-only research** skill. ## When to Use - User asks "what's the best library for X?" - User needs current documentation or API references - User wants to compare different libraries or approaches - User has an error message and needs solutions - User asks about best practices or security recommendations - User wants to learn about a new technology or framework - User needs to find tutorials or getting started guides ## Prerequisites Verify Codex CLI is available: ```bash codex --version # Should display installed version ``` **Requirements:** - Internet connectivity for web search - Authenticated Codex access (ChatGPT Plus/Pro/Team/Enterprise or API key in `~/.codex/config.toml`) - Web search enabled in `~/.codex/config.toml` as follows: ```toml [features] web_search_request = true ``` **Note:** The `--search` flag enables Codex's native web search tool. If web search is not available in your environment, the skill will provide information based on the model's training data. ## Basic Usage ### Step 1: Parse the Research Need Extract wh

What's inside
Steps it walks through
  1. When to Use
  2. Prerequisites
  3. Basic Usage
  4. Step 1: Parse the Research Need
  5. Step 2: Gather Local Context (Optional)
  6. Step 3: Execute Codex Search
  7. Step 4: Present Results
  8. Example Queries
  9. Documentation Lookup
  10. Library Comparison
  11. Error Resolution
  12. Best Practices
  13. Getting Started
  14. Output Format
Ships with 1 file
  • metadata.json
Commands it runs
codex --version  # Should display installed version
Check current dependencies
cat package.json  # or requirements.txt, go.mod, etc.
Check versions in use
npm list --depth=0  # or pip list, go list, etc.
codex --sandbox=read-only --search exec "Research and provide comprehensive information about: [QUERY]
Search context:
Do NOT make any changes - this is research only."
codex --sandbox=read-only --search exec "Find official documentation for React hooks in 2026:
Include only official React documentation."
More from claude-skill-registry
All skills →
About this skill
What does the codex-search skill do?

Search the web using OpenAI Codex CLI to find current documentation, best practices, solutions, and technical information. Use when the user needs to research libraries, find API documentation, troubleshoot errors, or learn about new technologies. Requires Codex CLI installed.

How do I install it?

Run `npx skills add majiayu000/claude-skill-registry --skill codex-search --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.

Where does this skill come from?

From majiayu000/claude-skill-registry, a repository with 534 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.

Is a popular skill a good skill?

Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.

Keep going