Tag
All content about Skills, organized for fast scanning.
Recent developments in the realm of skills and plugins highlight a push towards greater integration and efficiency in AI tools. Claude is launching a plugin portal to enhance submission and usage tracking, while OpenAI has introduced Agent Plugins to standardize agent skills across various platforms. Additionally, maintenance features are being implemented to optimize performance and reduce resource usage, reflecting a broader trend towards improving the functionality and usability of AI systems.
Claude is preparing a plugin portal for submissions, review tracking, and usage insights, based on a post shared by Boris Cherny quoting ClaudeDevs. The group says plugins bundle MCP and skills and claims MCP usage is up 110x this year, without data.
OpenAI has just rolled out Agent Plugins, an open standard to package Agent Skills and MCP server configs for use across supported clients. Launch partners include ChatGPT, Codex, Cursor, GitHub Copilot, VS Code, and Vercel—but Anthropic tools aren’t listed yet.
Claude Code is getting a new /checkup maintenance command aimed at tidying unused skills, MCPs, and plugins. It can also deduplicate and split CLAUDE.md files, disable slow hooks, and update settings to reduce per-session context overhead.
Antigravity has just rolled out Gemini 3.5 Flash (Low), aiming to use about 45% fewer tokens than the Medium setting while still topping Gemini 3 Flash (High) on SWE tasks. Product lead Varun Mohan also says Gemini quotas were reset for all plans after user feedback.
A recent Kilo Blog post by Brendan O’Leary explores what changes when a first designer joins a team where engineers and AI agents ship at “agent speed.” He argues design docs must become machine-actionable infrastructure to prevent product drift.
Superpowers 5.1.0 is out with a tighter, more compatible workflow, including revamped git worktree support for Claude Code and Codex. It also removes legacy slash commands, streamlines the code reviewer subagent, and improves OpenCode with new Factory Droid support.
Anthropic’s Boris Cherny shared new workflow tweaks for Claude Opus 4.7 focused on longer, less interruption-prone coding runs. Highlights include auto mode for safer auto-approvals, recaps and focus mode, and verification tools like /go for end-to-end tests.
Agents can be smart and still unreliable for real work when they lack domain expertise. This talk argues that the next step isn’t more agent scaffolding—it’s packaging reusable expertise as “Skills” that agents can load when needed. Key takeaways Breaks down what Skills are: organized folders of procedural knowledge (including scripts/tools) that can be versioned, shared, and composed. Explains why code and the file system can act as a universal interface, while Skills supply the missing domain context. Walks through “progressive disclosure” to protect the context window and enable libraries of hundreds or thousands of Skills. Maps an emerging architecture: agent loop + runtime environment + MCP servers for connectivity, with Skills providing expertise—and how that supports deployment across domains.