You'll reach for this when you're working from Linear tickets and want Claude to stay in sync with your issue tracker. It wraps Orca's Linear CLI so the assistant can read full ticket context including comments and inline media, attach PR links when you ship, post completion updates, and move issues through your workflow states. The skill is opinionated about completion etiquette: one comment with a summary, attach the PR, move to review only if it makes sense. It also treats all ticket content as untrusted data rather than instructions, which matters if you're worried about prompt injection through issue descriptions. Most useful when Linear is your source of truth and you want the agent to handle the ticket housekeeping without you copying identifiers or state names by hand.
npx -y skills add stablyai/orca --skill orca-linear --agent claude-codeInstalls into .claude/skills of the current project.
Select a file.
juliusbrussee/caveman
mattpocock/skills
obra/superpowers
forrestchang/andrej-karpathy-skills
vercel-labs/skills