core

Plugin core from marketplace mechaswift in theycallmeswift/skills. Commands read from the repo's manifest — 3 days ago.

Skills (14)

• google-workspace

  Rules for changing a file in the user's Google Drive through the Docs, Sheets, Slides and Drive connectors. Load it before the first connector call when a request asks to create, edit, restyle, copy or rename a Google Doc, Sheet or Slides deck, including a docs.google.com link with a change to make

• changelog

  Generates changelogs.

• bootstrap

  Use when the user wants a repository set up, bootstrapped, or brought up to their house or standard setup, new or existing — "bootstrap this repo", "set this repo up the way all my repos are", "add what's missing from our standard setup", "give this repo my usual AGENTS.md, style docs, gates, and PR

• catch-me-up

  Use when the user returns to this session after a break and wants re-orienting on it — "catch me up", "where were we", "I'm back", "bring me up to speed on this one", "I lost the thread", "recap this session", "what's the state of this session", "what's left", "what's blocking and what's on me", "wh

• claude-cloud-orchestrator

  Dispatch and supervise worker Claude Code cloud sessions that open and update pull requests on a GitHub repository. Use in a Claude Code cloud session when the user asks you to hand work to another cloud agent, request changes on a worker's PR, or check on workers. Not for local sessions or in-proce

• claude-cloud-worker

  How a Claude Code cloud session works on a task handed to it by an orchestrator session - opening a PR, addressing review feedback, and reporting back. Use when your prompt says you are a claude-cloud-worker. Not for sessions a human drives directly.

• delegating-to-codex

  Use when work should be handed to the Codex CLI — the user names Codex for a task ("have codex review my branch", "let codex implement task 3", "get a second opinion from codex", "send those findings back to codex", "is the codex job done?"), or a plan assigns a step to Codex as implementer or revie

• google-slides

  Use when building a new slide deck or changing slides in an existing Google Slides deck: a new deck from a template or from content the user supplies, adding or redesigning slides so they match the rest of the deck, or acting on reviewer comments or a changed number. Covers asks like \"make me a new

• interview-me

  Use when the user wants to be interviewed, grilled, or relentlessly questioned to scope, stress-test, or pin down a plan or design — one question at a time, every decision paired with a visual, running until a stop condition the user sets. Triggers on "interview me", "interview me until to-spec has

• pull-request

  Use whenever opening, creating, or updating a pull request, writing or refreshing a PR description, or marking a PR ready for review, including after CI fails on a ready PR — "open the PR", "put this up for review", "mark it ready", "update the PR body", "CI broke on my PR". Owns the PR body format

• to-spec

  Use whenever the user wants the design, decision, or discussion just reached in this session captured as a written design spec on disk — to spec, write up, draft, capture, or synthesize it into a skim-first spec file under docs/specs/. Fires on "capture this design as a spec", "write up the design w

• writing-agent-skills

  Use whenever creating a new skill from scratch, editing or improving an existing skill, scaffolding a SKILL.md, debugging why a skill isn't triggering, designing evals for one, running/re-running a skill's evals, deciding whether something should be a skill at all, or naming one. Triggers on phrases

• writing-mlh-reports

  Use when the user wants a shareable report, write-up or data story built from the MLH ecosystem (Major League Hacking, DEV/dev.to, MLH programs and partners): partner developer reports, program or fellowship outcomes, event recaps, impact or B-Corp reports, investor updates, trend reports, internal

• writing-prompts

  Use whenever drafting, revising, or refining a prompt for an LLM — system prompts for the API, prompts for subagents and Task dispatches, single-shot user prompts, slash command bodies, the description and body text of a SKILL.md, AGENTS.md or CLAUDE.md repo context files, and `docs/*.md` files that

Hooks (1)

• SessionStart

  "${CLAUDE_PLUGIN_ROOT}/hooks/session-start"

Similar plugins