claudecodex

u

A grab-bag of useful commands for various tasks

1 on GitHubby Tom Ashworthv0.20.1MITupdated 9 days agoDeveloper Tools
/

Plugin u from marketplace tgvashworth-agent-plugins in tgvashworth/agent-plugins. Commands read from the repo's manifest — 9 days ago.

What it installs

Skills (18)

  • bg

    Run a task in a background agent. Use when the user invokes /bg with a task description to spin off an autonomous sub-agent.

  • commit-push-draft

    Commit the current changes, push the branch, and open a draft pull request, or update an existing PR in place without changing its draft state. Use when the user invokes /commit-push-draft, asks for a draft PR, or wants to push work up for a look before it is ready for review.

  • commit-push-pr

    Commit the current changes, push the branch, and open a pull request ready for review, or update an existing PR and mark a draft ready. Use when the user invokes /commit-push-pr, says "ship this", "open a PR", or asks to commit and push and create a pull request.

  • commit

    Create a single git commit from the current changes, with a message in the repository's commit style. Use when the user invokes /commit, says "commit this", or asks for a commit without a push or pull request.

  • context

    Get up to speed on a task: inspect the branch, working tree, recent commits, and any ticket, issue, or PR the user points at, then summarise what is being worked on and what to do next. Use when the user invokes /context, switches to a new piece of work, or asks to get oriented or catch up on a bran

  • deslop

    Rewrite words for one-pass comprehension: code comments, user-facing copy and microcopy, commit messages, PR titles and bodies, ticket titles and bodies. Plain English, short sentences, no filler, no AI tells, no personification, no agent working memory. Use when the user says "deslop", "fix the cop

  • pr-comments

    Fetch all PR comments, reviews, and review threads in a single GraphQL call. Use when you need to check PR feedback, address review comments, see review status, or understand what reviewers are asking for on a pull request.

  • pr-feedback

    Review PR feedback from reviewers and Copilot, triage each comment (implement / acknowledge / decline), plan changes, and implement approved fixes. Use when you need to act on pull request review feedback.

  • pr-respond

    Respond to and resolve PR review comments. Reply to reviewer feedback, resolve addressed threads, and post follow-up comments. Use after implementing changes from PR feedback or when closing the loop with reviewers.

  • quiz

    Test the user's understanding of a change or pull request with a fast, adaptive, evidence-backed quiz tied to the actual diff. Use when the user asks to quiz them, test their understanding, check that they understand their change, or find gaps in their mental model of code they just changed.

  • ready-for-review

    Rewrite an existing draft PR's title and body from scratch, then mark it ready for review. Use when the user says "ready for review", "this draft is ready", "open my draft PR for review", or asks to tidy up a draft PR's description before sending it to reviewers.

  • remember

    Append a convention, fact, or instruction to the project's shared AGENTS.md or CLAUDE.md so it is committed to the repo and shared with the team. Use when the user wants to persist project knowledge beyond auto memory, when they say "remember this", or when a significant convention or correction com

  • review-agent

    Kick off a background review of a PR and watch it for incoming feedback. Spins up an agent that runs the built-in /review skill against the current (or a specified) PR, then starts a persistent watcher that streams new CI failures and review comments (bots and humans) as they land — triaging everyth

  • review-cycle

    Drive a pull request through review: investigate feedback, fix issues, test, commit, push, and respond until CI is green and actionable comments are addressed, with a rising stop-here score to prevent endless iteration. An independent review sub-agent is optional. Use when the user says "review-cycl

  • slack

    Write a Slack message from the current context, in the formatting dialect that will actually render for how it is delivered: returned as bare text to copy into the composer (default), saved as a Slack draft, or sent. Use when the user invokes /slack, or asks for something they can paste into Slack,

  • tourguide

    Walk through code changes or a plan in a logical, grouped sequence — one group at a time, stopping after each to wait for the user to say "next" (or give feedback) before continuing. Instead of showing a raw diff or dumping a plan all at once, organise related pieces together and present them increm

  • what

    Re-explain your previous message in plain English, reconnecting it to the conversation the reader has only lightly followed. Use when the user invokes /what, or says they don't understand what you just said and wants it rewritten more clearly.

  • write

    Write or improve a substantial document through a structured process: fix the goals, outline the key points, map the audience, design the structure, draft it, then run two independent review passes (a cold-read for clarity and a style pass for LLM-isms and British English). Use when the user wants t

View source on GitHub

More from tgvashworth/agent-plugins

All 2 plugins →

playbook-dev

claudecodex

Create structured, multi-stage analysis playbooks for LLM-driven workflows

22 skills · 1 agent

1

Similar plugins

superpowers

claudecodex

Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques

15 skills · hooks

obra/superpowers 316,836

CMEM Pro

claude

CMEM Pro for the Claude app - hooks stream tool use to cmem.ai and inject observations into new sessions and agents

3 skills · 1 MCP server · hooks

thedotmack/claude-mem 116,618

context7

claudecodex

Up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your coding agent's context.

1 MCP server

upstash/context7 64,709

mempalace

claudecodex

AI memory system — mine projects and conversations into a searchable palace. 47 MCP tools, auto-save hooks, guided setup.

3 skills · 6 commands · 1 MCP server · hooks

MemPalace/mempalace 60,536

Build powerful, stateful CLI interfaces for any GUI application using the cli-anything harness methodology.

5 commands

HKUDS/CLI-Anything 58,665

Reliable automation, in-depth debugging, and performance analysis in Chrome using Chrome DevTools and Puppeteer

8 skills · 1 MCP server

ChromeDevTools/chrome-devtools-mcp 55,181

Maintain u?

This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.

claude & codex plugins2 plugins