Productivity plugins

Cross-model orchestrator: pins frontier primary models while routing subagents to haiku scouts and sonnet workers with lifecycle guardrails.

1 skill

Raw idea in, evidence-backed build blueprint out. Research, architecture, pinned stack, epics, stories, acceptance criteria and a one-epic-per-session build protocol.

1 skill

Side agents watch longer tasks and pass what they find to the main agent automatically. Add your own with a custom prompt, and see how each note landed.

1 skill

Reading Scout is a Codex plugin that finds papers, essays, and research posts, checks the original sources, and explains why each pick fits. It can save the reads you choose to Notion or a Space Page.

1 skill

Turn AI conversations into illustrated learning notes and save them to ZNote.

1 skill

End a session on purpose: consolidate what it learned into current-state notes, then clean up only what it left behind — leftover watchers, schedules, temp files — and report what is still open. Lives in its own repo: https://github.com/Miguel-Barroso/claude-wrap-up

1 skill

Sets up, audits and tunes a coding agent for DS/ML work: short protected core rules, a domain profile, done-means-checked, one-phrase tune.

1 skill

One command from a GitHub issue, Jira ticket or bug description to a tested fix and a draft pull request.

1 command

Converts a code repository into a single readable document that can be uploaded to NotebookLM for AI-powered podcast generation and Q&A.

1 skill

Bootstrap a freshly cloned repository: read docs, detect the stack, install dependencies, configure environment, auto-generate non-secret values, and verify the project runs.

1 skill

毎朝、開発ニュースのフィードを集めて Slack に届ける

1 skill

Slack のフォーム投稿から GitHub Issue を起票する（Orca automation の precheck で、投稿があるときだけ Claude を起動）

1 skill

Hooks that load TaskGraph state at session start, keep the task in progress in view, and keep Claude going while its acceptance criteria are unchecked. Use with the taskgraph plugin.

hooks

AI-powered development tools. 29 agents, 22 commands, 19 skills, 1 MCP server for code review, research, design, and workflow automation.

Pins the Vercel deploy queue of the linked project under the prompt: every deploy queued, building or just finished, with its phase and elapsed time

Shared knowledge definitions and workflows for people and agents

Persistent memory, knowledge, and skill workflows for coding agents.

Catalogo local das skills deste repositorio para uso no Codex.

Starter pack for VibeCoding

No description.

cBranch plugin

Large previews of pasted images above the Claude Code prompt, with captioned thumbnails while you type

Warns when a turn bloats the context or auto-compaction is near, and /footprint shows what fills the context and which session holds the RAM

Claude Code's prompt-cache state, drawn above the prompt where no statusline renders: the Desktop app's Code tab. Each response graded as `warmline audit` grades it, from its real cache-read and cache-write counts.

Your Claude Code 5-hour and 7-day usage limits as animated bars under the prompt: percent, reset countdown, pace mark, projection, per-answer delta, styles (Pac-Man, beer, tank, battery, hourglass)

A collection of reusable skills for Codex, Claude Code, and other AI agents. Built by Long

Dynamically provisions AI agent skills based on project tech stack detection

TokenMonitor inside Claude Code, no device needed. A band above the prompt shows your Claude 5h/7d quota bars, reset countdowns, the projection to reset, context fill and session cost, plus your Codex and Antigravity quotas when those CLIs are signed in on this machine (read-only use of their sign-ins; each one can be switched off). /token-monitor-hud opens a pane with burn rate, pace and time-to-100 %; toasts at 80 % and 95 %. Claude Code only.

No description.

See pictures in the transcript, as the desktop app shows them: the images Claude reads, the ones you paste and those its answers link to; GIFs play.