Agent skills for Zed

Write and edit project docs (README/markdown) as a glossary of principles, not a mirror of the code.

Generate documentation for custom elements based on their manifest data.

Summarize the current session into a lean, AI-only handoff doc so the next agent can resume cold in a fresh context.

Author a declarative invariant rule module for `jidoka invariants`.

Reach for the smallest solution that actually works before writing code: question whether the task needs to exist (YAGNI), prefer the…

Write, rewrite, review, or continuously refine X/Twitter posts in Meng To's current voice using his deduplicated authored-post corpus…

Use when creating or modifying an OpenGL shader or its integration.

Write or develop an opinion piece (opiniestuk/op-ed).

Use whenever a task writes or modifies PHP code, including PHP files, embedded PHP, tests, scripts, endpoints, and PHP code snippets.

Picks the right writing register per output segment, automatically.

Generate and compare exactly three code scaffolds through a write-review-execute loop, then recommend the best solution using weighted…

How to write modern Swift well — modeling with value types, Swift 6 data-race safety and approachable concurrency (@concurrent…

Write tests that pin real behavior instead of implementation details, config values, or lucky samples.

Use this skill when writing, drafting, or improving a feature ticket for the Looq.ai Linear workspace.

Prose quality checker for Quarto (.qmd) files, grounded in William Zinsser''s *On Writing Well* (30th Anniversary Edition).

Use when working on substantial writing artifacts where structure, audience fit, critique, revision, translation, style learning, or…

Guides implementation of Gatekeeper Workers that bridge Gadgets to external services.

How to write great agent instructions for an agent-native app or template: AGENTS.md, skills, and tool/action descriptions.

Writing and maintaining GitHub issues for the tldraw repository.

Write in John Wiegley's authentic voice.

Gates whether a code comment should exist and forces the ones that stay to explain why, not what.