Devotd Github
NewAI agents

agentskillexchange/skills — regenerated with 3,200 catalog entries

agentskillexchange/skills automatically regenerated the entire repository, raising the number of published skills to 3,200 and adding a new /call-me skill.

Repositoryagentskillexchange/skills ↗

agentskillexchange/skills automatically regenerated the entire repository, raising the number of published skills to 3,200 and adding a new /call-me skill.

GitHub description: Curated, trusted open catalog of AI agent skills for OpenClaw, Claude Code, Codex, GitHub Copilot, Gemini, Cursor, MCP, LangChain, and more. Browse, install, and verify reusable agent workflows.

stars
★ 52
Forks
85
Language
Python
License
MIT
Read original source ↗View on GitHub ↗

agentskillexchange/skills is a public catalog that collects, verifies, and distributes skills for AI agents (procedure files that organize repetitive tasks). The commit 111d7f3 of 2026-11-07 07:27 UTC, with the message "chore: regenerate all repo files from live data", automatically regenerated the entire repository, adding 64 lines across 85 files and deleting 21 lines.

What actually changed with the regeneration was the catalog numbers. Published skills rose from 3,199 to 3,200, while the 17 categories and 15 industry collections stayed the same, and the security-review figure remained unchanged at 2,621. The only newly registered item is /call-me. It is a skill that lets an agent place calls or send text messages to the user's own iPhone or Android device via a remote MCP server, so the Integrations & Connectors category grew from 189 to 190. The README badge and the description strings in .claude-plugin, .codex, .cursor-plugin, and .opencode also changed from "3,100+" to "3,200+".

That said, this commit is not code development but data regeneration. The catalog's recorded time moved from 2026-11-07 01:26 UTC to 07:27 UTC, and in that interval published skills grew by only 1, from 3,199 to 3,200. The new entry /call-me is recorded only with the status "Published" and is not included among the security-reviewed items. The plugin description characterizes the whole set as security-scanned, but the catalog counts 2,621 security reviews, and no baseline was given for comparison by individual skill star counts or download counts. The listing is published under the MIT license, but since it contains sensitive behavior such as a skill in which an agent places calls and sends text messages through the user's phone, it needs to be checked before being used for any purpose.