Search skill packs and skills together

AI-built interfaces can drift into inconsistent patterns and unclear design choices. This skill gives agents focused constraints for interface work.

React and Next.js apps often lose speed through waterfalls, large bundles, and avoidable re-renders. This skill gives agents focused rules for reviewing and improving performance-sensitive code.

Agents can claim success before checking real evidence. This skill requires fresh verification commands and evidence before completion claims.

Voice agent teams must balance latency, turn taking, speech quality, and control. This skill guides architecture choices and production design tradeoffs.

Large agent instruction files waste context and hide important rules. This skill splits them into a compact root file and focused linked documents.

Messy staging and weak messages slow code review. This skill guides selective staging, commit splitting, Conventional Commits messages, and verification.

Teams can lose requirements when agent-skill work moves from planning to implementation and review. This skill keeps requirements, acceptance evidence, review state, and completion claims connected through a clear traceability process.

SpecDD projects can be hard to interpret when rules are spread across bootstrap files and specs. This skill explains the active contract in plain language without editing files.

SpecDD repositories can be hard to enter without knowing the active bootstrap and authority model. This skill guides Claude, Codex, and Claude Code through a read-only orientation pass before later work.

AI agents need repeatable browser actions that provide evidence, not fragile manual steps. This skill guides Claude, Codex, and Claude Code through deterministic navigation, snapshots, screenshots, and form workflows.

Release updates are hard to track when changes are scattered across tickets and commits. This skill organizes changelogs, release notes, and migration guidance into clear user-facing documents.

Teams often overbuild while trying to make failing tests pass. This skill guides Claude, Codex, and Claude Code toward minimal green-phase implementation.