Codex
Codex generates clean, semantic HTML from natural language descriptions, streamlining frontend development and boosting accessibility.
Install on your platform
We auto-selected Claude Code based on this skill’s supported platforms.
Run in terminal (recommended)
claude mcp add davila7-codex npx -- -y @trustedskills/davila7-codex
Or manually add to ~/.claude/settings.json
{
"mcpServers": {
"davila7-codex": {
"command": "npx",
"args": [
"-y",
"@trustedskills/davila7-codex"
]
}
}
}Requires Claude Code (claude CLI). Run claude --version to verify your install.
About This Skill
What it does
The Codex skill generates clean, semantic HTML code based on natural language descriptions. It streamlines frontend development by translating user instructions into structured HTML markup. The skill defaults to using the gpt-5.2 model and allows for configuration of reasoning effort and sandbox mode to control access and safety.
When to use it
- Generating basic HTML structures from text descriptions (e.g., "Create a header with a title").
- Reviewing existing HTML code in a read-only mode.
- Applying local edits or modifications to HTML files.
- Resuming previous Codex sessions for continued development or analysis.
Key capabilities
- Generates semantic HTML from natural language instructions.
- Supports configuration of reasoning effort (
xhigh,high,medium, orlow). - Offers different sandbox modes:
read-only,workspace-write, anddanger-full-access. - Allows resuming previous sessions via stdin input.
- Suppresses thinking tokens (stderr) by default for cleaner output.
Example prompts
- "Create a simple HTML page with a heading that says 'Welcome'."
- "Generate an unordered list of fruits."
- "Resume the last Codex session."
Tips & gotchas
- By default, Codex suppresses thinking tokens; request to see them only when debugging or explicitly needed.
- Always use
--skip-git-repo-checkwhen running Codex commands. - When resuming a previous session, avoid specifying configuration flags unless specifically requested by the user.
Tags
TrustedSkills Verification
Unlike other registries that point to live repositories, TrustedSkills pins every skill to a verified commit hash. This protects you from malicious updates — what you install today is exactly what was reviewed and verified.
Security Audits
| Gen Agent Trust Hub | Pass |
| Socket | Pass |
| Snyk | Pass |
🌐 Community
Passed automated security scans.