Anthropic Proposes Function Hooks to Make Claude Code Plugins '10x More Powerful'
poteat · ghdev · 2026-09-04
Anthropic engineers have opened a community proposal for Claude Code called Function Hooks: TypeScript functions that can deeply modify Claude Code behavior, kept safe via side-effect tracking on a parameterized $ object and composable through an Express/Koa-style registration-order 'next' continuation model.
Key points:
- Workspace admins could programmatically fine-tune what changes users can make
- Component rendering itself could be modified, leveraging the React-based codebase
- An architecture doc plus multiple 60-second demo videos are attached
- The team says community engagement will decide whether this ships; the $ sigil is non-negotiable
More from coding & agent
- LukeW built an agent-orchestration app for a foldable: managing agents is now an all-day job — LukeW · 2026-10-21
- Agentic SDLC: engineers shift from implementers to agent orchestrators as cycles shrink from weeks to hours — Pavan_Belagatti · 2026-10-21
- OpenAI's dots can now use company tools via Merge for Workforce with employee-bound permissions — shensi · 2026-10-21
- The topology of an LLM agent: where the model, loop, and tools run shapes cost and security — davidcrawshaw · 2026-10-21
- Agents with full-disk and accessibility access are an enterprise security nightmare, argues Runlayer — leveragedupside · 2026-10-21
- Acellera runs an unsupervised virtual lab of AI scientists on a real IL-23 molecular discovery program — gdefabritiis · 2026-10-21