Audit AI skill safety before you ship. Static + semantic + adversarial audit pipeline for SKILL.md, Codex skill.yaml, Claude Code commands, and OpenCode. Free hosted tier on tarai.dev.
Skill Claude CodeCodex
Echo the user's input back verbatim. Useful for greetings, smoke tests, or any wish that's purely conversational and doesn't need data fetching or analysis. Activates when the user says "echo X", "hi", "hello", or similar.
Skill Claude CodeCodex
Fetch a URL and produce a concise one-paragraph summary. Activates when the user wish contains a URL or asks to "summarize", "read", "fetch", or "tell me what this page says about" a webpage.
Skill Claude CodeCodex
Generate draft text for one or more social platforms from a topic or URL. Wraps postall-agent generate. Activates when the user wish includes a content topic plus a publishing intent ("write a tweet about X", "draft a LinkedIn post on Y", "summarize this URL into a wechat article").
Skill Claude CodeCodex
Create a cover image for an existing draft. Wraps postall-agent image. Activates when the user wish references a draft and asks for "a cover image", "add a picture", or comes right after a generate-content step in a publishing pipeline.
Skill Claude CodeCodex
List recent content drafts. Wraps postall-agent list. Activates when the user wish is read-only — "what drafts do I have", "show me unpublished posts", "any pending content".
Skill Claude CodeCodex
Publish an existing draft to one or more platform accounts. Wraps postall-agent publish. Activates when the user wish says "publish", "send it out", "post to LinkedIn", or comes at the end of a publishing pipeline.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: