Skip to content
Back to search

skills

Agent skills for Claude Code and other AI agents

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today

Check it deeper on your machine

It has changed since its scan on 11 Nov 2026.

As a Claude Code plugin, install it and then ask Claude Code:

/plugin marketplace add keithmackay/mackayi/plugin install scanskillsafety@mackayiDo a deep scan of https://github.com/swyxio/skills with scanskillsafety

Or from a terminal, clone it and run it:

git clone https://github.com/keithmackay/scanskillsafety-skill ~/.claude/skills/scanskillsafetynode ~/.claude/skills/scanskillsafety/cli.cjs --deep 'https://github.com/swyxio/skills'

It checks the exact commit: every file, its packages against OSV (api.osv.dev), how it compares with this page's scan (findsafeskills.com receives the repository's address), and, if you agree, Claude reading every file with your own Claude subscription. It never runs the code.

Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/swyxio--skills/badge.svg)](/listings/swyxio--skills)

HTML

<a href="/listings/swyxio--skills"><img src="/api/listings/swyxio--skills/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

Clone the collection with Git, then link a selected skill into your agent's personal skills directory. You need Git and a compatible agent installed. This Codex example uses macOS/Linux and HTTPS, so cloning does not require a GitHub SSH key:

git clone https://github.com/swyxio/skills.git ~/Work/skills
mkdir -p ~/.agents/skills
test ! -e ~/.agents/skills/review-thread && test ! -L ~/.agents/skills/review-thread && \
  ln -s ~/Work/skills/review-thread ~/.agents/skills/review-thread
test -r ~/.agents/skills/review-thread/SKILL.md && echo "review-thread linked"

The last command prints `review-thread linked` when the installed instructions are readable. The checks before `ln -s` prevent replacing an existing entry: inspect that entry before deliberately replacing it. Readability alone does not prove that an existing entry points to this checkout.

Choose the destination for your agent:

| Agent | Personal skills directory | Use | | --- | --- | --- | | [Codex](https://developers.openai.com/codex/skills) | `~/.agents/skills/` | Select the skill or mention `$review-thread` in a prompt. | | [Claude Code](https://code.claude.com/docs/en/skills) | `~/.claude/skills/` | Link there instead, then invoke `/review-thread`. | | [Cursor](https://cursor.com/docs/skills) | `~/.cursor/skills/` or `~/.agents/skills/` | Link there, then select the skill in Agent chat. |

For a first use, open a task that already contains your goal and work so far (or provide that context), then ask: "Use review-thread to assess progress toward my goal, what remains, and the likely agent execution time." Expect an assessment grounded in that task's context; it does not start new tasks merely by suggesting them. A useful assessment distinguishes what works from what is merely implemented or tested, and names the remaining completion endpoint. Refresh or restart the agent if the skill is missing from its selector. The author also maintains a legacy `~/.codex/skills/` installation; the table follows current documented directories for new installs.

Repeat the link command for the skills you need. Keep a single checkout as the source and link whole skill folders so references and scripts stay together. Updates to that checkout appear through its links; use `git status` before `git pull --ff-only`, and resolve local edits rather than overwriting them. Link newly added skills separately. Installation supplies instructions; it does not install workflow dependencies or grant API, browser, or publishing access.

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 383 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.

  1. # swyxio Skills
  2. Reusable skills for Claude Code, Codex, Cursor, and similar agent environments.
  3. Each skill folder contains a `SKILL.md`: a trigger description followed by the
  4. workflow an agent reads when the skill applies. Scripts, references, and assets
  5. support individual workflows. This is a personal working collection; some
  6. skills assume swyx's projects, preferences, credentials, or installed tools.
  7. Read a skill's prerequisites before running its commands.
  8. ## Install And Try One Skill
  9. Clone the collection with Git, then link a selected skill into your agent's
  10. personal skills directory. You need Git and a compatible agent installed. This
  11. Codex example uses macOS/Linux and HTTPS, so cloning does not require a GitHub
  12. SSH key:
  13. ```bash
  14. git clone https://github.com/swyxio/skills.git ~/Work/skills
  15. mkdir -p ~/.agents/skills
  16. test ! -e ~/.agents/skills/review-thread && test ! -L ~/.agents/skills/review-thread && \
  17. ln -s ~/Work/skills/review-thread ~/.agents/skills/review-thread
  18. test -r ~/.agents/skills/review-thread/SKILL.md && echo "review-thread linked"
  19. ```
  20. The last command prints `review-thread linked` when the installed instructions
  21. are readable. The checks before `ln -s` prevent replacing an existing entry:
  22. inspect that entry before deliberately replacing it. Readability alone does not
  23. prove that an existing entry points to this checkout.
  24. Choose the destination for your agent:
  25. | Agent | Personal skills directory | Use |
  26. | --- | --- | --- |
  27. | [Codex](https://developers.openai.com/codex/skills) | `~/.agents/skills/` | Select the skill or mention `$review-thread` in a prompt. |
  28. | [Claude Code](https://code.claude.com/docs/en/skills) | `~/.claude/skills/` | Link there instead, then invoke `/review-thread`. |
  29. | [Cursor](https://cursor.com/docs/skills) | `~/.cursor/skills/` or `~/.agents/skills/` | Link there, then select the skill in Agent chat. |
  30. For a first use, open a task that already contains your goal and work so far
  31. (or provide that context), then ask: "Use review-thread to
  32. assess progress toward my goal, what remains, and the likely agent execution
  33. time." Expect an assessment grounded in that task's context; it does not start
  34. new tasks merely by suggesting them. A useful assessment distinguishes what
  35. works from what is merely implemented or tested, and names the remaining
  36. completion endpoint. Refresh or restart the agent if the skill is missing from
  37. its selector. The author also maintains a legacy `~/.codex/skills/` installation;
  38. the table follows current documented directories for new installs.
  39. Repeat the link command for the skills you need. Keep a single checkout as the
  40. source and link whole skill folders so references and scripts stay together.
  41. Updates to that checkout appear through its links; use `git status` before
  42. `git pull --ff-only`, and resolve local edits rather than overwriting them.
  43. Link newly added skills separately. Installation supplies instructions; it does
  44. not install workflow dependencies or grant API, browser, or publishing access.
  45. ## Quick Routing
  46. Use the most specific skill that matches the job. When a workflow spans multiple stages, start with the orchestrator skill, which coordinates the stages, then hand off to the atomic
  47. skill that handles one stage.
  48. | User intent | Start with | Hand off to |
  49. | --- | --- | --- |
  50. | Plan or resume an annotated episode at any stage; own overlay selection, treatment, and encoded-result review | [annotated-video-production](./annotated-video-production) | Atomic capture, transcript, thumbnail, and YouTube skills; media-transform owns generic transfer pipelines |
  51. | Download, transform, transcribe, thumbnail, or publish media end-to-end | [media-transform](./media-transform) | `download-*`, `transcribe-anything`, `thumbnail-extraction`, `youtube-*` |
  52. | Download a video from a page, X/Twitter, or Zoom | [download-video](./download-video), [download-x-video](./download-x-video), or [zoom-download](./zoom-download) | [media-transform](./media-transform) if more stages follow |
  53. | Upload many submitted talks from Airtable/local files to YouTube Studio | [youtube-studio-batch-upload](./youtube-studio-batch-upload) | [youtube-studio-computer-use](./youtube-studio-computer-use) for post-upload Studio cleanup |
  54. | Edit existing YouTube Studio videos, thumbnails, playlists, visibility, or schedules through Chrome | [youtube-studio-computer-use](./youtube-studio-computer-use) | [youtube-api](./youtube-api) if API credentials exist and the task is API-friendly |
  55. | Use the YouTube Data API for metadata, thumbnails, uploads, or channel listing | [youtube-api](./youtube-api) | [youtube-studio-computer-use](./youtube-studio-computer-use) for Studio-only states |
  56. | Build a full YouTube operations bot with raw API access, Slack approvals, playlists, comments, live operations, and change-impact analytics | [youtube-channel-operator](./youtube-channel-operator) | [slackbot-builder](./slackbot-builder), [data-chatbots](./data-chatbots), then [youtube-api](./youtube-api) or [youtube-studio-computer-use](./youtube-studio-computer-use) |
  57. | Redesign an app or explore product behavior through generated visual directions and matched implementation screenshots | [design-apps-with-imagegen](./design-apps-with-imagegen) | [visual-playtest](./visual-playtest) after the selected direction is implemented |
  58. | Visually inspect a local or deployed site/app and find concrete layout or responsive defects | [visual-playtest](./visual-playtest) | Load its app/media references only when those workflows are in scope; use [design-apps-with-imagegen](./design-apps-with-imagegen) for broader redesign |
  59. | Create or align a durable project CEO, product steward, or autonomous owner | [ceo-creator](./ceo-creator) | Project-specific execution skills selected by the approved CEO charter |
  60. | Create a durable independent dissent agent with its own evidence model, ledger, interruption threshold, and cadence | [cassandra-creator](./cassandra-creator) | The created Cassandra thread's bounded tests or decisions |
  61. | Critically audit or trim an overgrown skill or over-broad trigger | [skill-cutter](./skill-cutter) | Apply local cuts only when explicitly requested |
  62. | Build conference schedule, speaker, or developer data surfaces | [schedule-design](./schedule-design), [conference-developer-endpoints](./conference-developer-endpoints), or [europe-developer-api](./europe-developer-api) | `accelevents-*` or [sync-accelevents](./sync-accelevents) when syncing source systems |
  63. | Harden a software repo | [antislop-codebase](./antislop-codebase) for deliberate structural cleanup | `productionize-*`, `security-*`, `observability-*`, `release-*`, `test-*` |
  64. | Design, debug, migrate, cache, or deploy a production system on Cloudflare | [cloudflare-production-builder](./cloudflare-production-builder) | Product-specific skills after the Cloudflare durability, storage, security, and release boundaries are settled |
  65. | Host repositories or deploy exact-SHA releases through SmolForge | [forge](./forge) | [cloudflare-production-builder](./cloudflare-production-builder) when the underlying Cloudflare runtime or bindings also need design work |
  66. | Build a structured-data chatbot or Slack bot | [data-chatbots](./data-chatbots) or [slackbot-builder](./slackbot-builder) | [app-ux-paradigms](./app-ux-paradigms) for interaction details |
  67. | Protect usernames and public handles from route collisions, squatting, or impersonation | [reserved-handle-policy](./reserved-handle-policy) | [security-hardening](./security-hardening) when broader auth or permission review is needed |
  68. | Create or substantially revise a repository README around a verified first result | [ai-readme](./ai-readme) | [ai-devblog](./ai-devblog) for a dated engineering story |
  69. | Write, revise, verify, or publish an individual technical devblog | [ai-devblog](./ai-devblog) | [blog-system-design](./blog-system-design) only when shared presentation infrastructure must change |
  70. | Create or redesign a technical blog index, post shell, navigation, search, typography, or reusable components | [blog-system-design](./blog-system-design) | [ai-devblog](./ai-devblog) for individual article content |
  71. ### Routing Notes
  72. - Prefer API skills for stable, supported bulk operations; prefer Computer Use skills for authenticated browser states, Studio-only controls, file pickers, disabled buttons, and save verification.
  73. - Keep source acquisition, metadata staging, and upload ledgers in batch/upload skills. Keep existing-video cleanup, thumbnails, scheduling, playlist fixes, and save recovery in `youtube-studio-computer-use`.
  74. - Treat status, reviewer, and operations fields as private by default. Public descriptions should use submitted abstracts, bios, company/project links, and social links entered for publication.
  75. - Do not physically reorganize skill folders into categories unless every target agent loader supports nested skill discovery. The top-level `folder/SKILL.md` layout is intentional.
  76. ## Skill Index
  77. ### Coding, Agents, And Workstations
  78. #### Kakuna Codebase Hardening Suite
  79. Use these opt-in skills only when the named hardening problem is the primary
  80. task. They are diagnostic tools, not a maturity ladder: select the smallest
  81. relevant skill, reuse existing controls, prefer deletion, and stop when the
  82. explicit problem is resolved. Ordinary implementation work should not trigger
  83. the suite merely because it will ship or could be made more robust.
  84. <table>
  85. <tr>
  86. <td width="280" align="center" valign="middle">
  87. <img src="./assets/kakuna-codebase-hardening.png" alt="Kakuna Codebase Hardening Suite logo: a cute armored cocoon mascot inside a code shield" width="240">
  88. </td>
  89. <td valign="middle">
  90. <p><strong>Foundation</strong></p>
  91. <ul>
  92. <li><a href="./antislop-codebase">antislop-codebase</a> — <strong>Structural cleanup/migration.</strong> Reduces demonstrated repository maintenance cost without imposing folder, file-size, compatibility, testing, or audit-artifact targets.</li>
  93. </ul>
  94. <p><strong>Productization</strong></p>
  95. <ul>
  96. <li><a href="./productionize-app-with-services">productionize-app-with-services</a> — <strong>Bounded productization.</strong> Adds only explicitly needed product services for named users and operators; it does not install a default SaaS maturity stack.</li>
  97. </ul>
  98. <p><strong>Safety</strong></p>
  99. <ul>
  100. <li><a href="./security-hardening">security-hardening</a> — <strong>Threat-scoped appsec.</strong> Fixes concrete reachable risks in an explicitly reviewed threat surface and reuses existing framework/provider defenses.</li>
  101. </ul>
  102. <p><strong>Operability</strong></p>
  103. <ul>
  104. <li><a href="./observability-hardening">observability-hardening</a> — <strong>Question-driven visibility.</strong> Uses the cheapest existing or new privacy-safe signal to answer named production questions without requiring every telemetry type.</li>
  105. <li><a href="./release-readiness-hardening">release-readiness-hardening</a> — <strong>Minimal ship proof.</strong> Audits the smallest sufficient release controls, prefers authoritative provider facts, and removes redundant ceremony.</li>
  106. <li><a href="./vercel-production-cost-review">vercel-production-cost-review</a> — <strong>Material cost diagnosis.</strong> Answers a defined Vercel billing question by investigating the few dominant drivers before authorizing any remediation.</li>
  107. </ul>
  108. <p><strong>Quality</strong></p>
  109. <ul>
  110. <li><a href="./test-strategy-hardening">test-strategy-hardening</a> — <strong>Confidence-per-minute.</strong> Solves explicit test-system problems through faithful boundaries, deterministic fixes, and deletion or consolidation—not test-layer growth.</li>
  111. </ul>
  112. </td>
  113. </tr>
  114. </table>
  115. #### Other coding/workstation skills
  116. - [review-thread](./review-thread) — reviews task progress and goal alignment, estimates agent execution time, and suggests worthwhile parallel work.
  117. - [align-me](./align-me) — pauses before a long autonomous run to surface material ambiguities as numbered, lettered choices with concrete tradeoffs, recommendations, and an `approve all` path.
  118. - [new-mac-setup](./new-mac-setup) — opinionated Apple Silicon Mac bootstrap for fullstack and AI work. Installs Homebrew, shell tooling, editors, AI tools, terminal setup, and macOS defaults in a repeatable run order.
  119. - [cloudflare](./cloudflare) — routes ambiguous or cross-product Cloudflare work to the relevant product skill.
  120. - [cloudflare-production-builder](./cloudflare-production-builder) — chooses among Workers, Pages, Workflows, Queues, Durable Objects, D1, R2, KV, Cache API, alarms, and Cron; then applies durable handoffs, safe caching, migration discipline, multi-tenant boundaries, observability, and live production verification.
  121. - [forge](./forge) — operates SmolForge as a Git repository and collaboration host or as an exact-SHA Deploy/Sites release control plane, while preserving explicit authentication, provider, preview, and production boundaries.
  122. - [claude-session-introspect](./claude-session-introspect) — inspects Claude Code session JSONL files at `~/.claude/projects/` for token totals, prompt counts, assistant turns, tool calls, compaction boundaries, and compaction summaries.
  123. - [deep-trajectory-analysis](./deep-trajectory-analysis) — reconstructs paired agent, game, or policy trajectories from exact shared pre-states, connects aggregate effects to first-divergence evidence, and validates machine-readable causal reports before promotion decisions.
  124. - [design-apps-with-imagegen](./design-apps-with-imagegen) — audits an existing interface, researches free Mobbin references, generates four distinct visual and behavioral directions including a wildcard, honors a selected or delegated choice, and implements and compares matched screenshots at relevant sizes.
  125. - [visual-playtest](./visual-playtest) — runs a proportional browser review of representative visual states; app-interaction and media-workflow checklists are selectively loaded only when relevant.
  126. - [ceo-creator](./ceo-creator) — creates a durable project CEO with an evidence model, authority boundaries, operating cadence, initiative and delegation rules, privacy protections, and a decision-oriented reporting contract.
  127. - [cassandra-creator](./cassandra-creator) — creates a durable independent dissent agent with separate evidence access, a dated assumptions and predictions ledger, symmetric skepticism, a high interruption threshold, explicit self-correction, read-only authority, and an optional approved recurring cadence.
  128. - [skill-cutter](./skill-cutter) — critically classifies and trims overgrown skills to their behavioral core, narrows over-broad trigger metadata, and separates provider constraints from optional or project-specific policy.
  129. - [smart-entity-resolution](./smart-entity-resolution) — resolves named people or organizations in messy databases with aliases, duplicates, sparse records, common names, LLM retrieval repair, reranking, and visible runner-up candidates.
  130. - [autoreview](./autoreview) — performs an explicitly requested final review using whatever review capability is available, without depending on a particular helper, model, or service.
  131. - [public-qa-chatbot](./public-qa-chatbot) — builds unauthenticated public Q&A chatbot widgets with rate limits, origin/input hardening, semantic caching, observability, streaming UX, and robust chat scroll behavior.
  132. - [slackbot-builder](./slackbot-builder) — builds production Slack bots with signed Events API handlers, causal shared-thread sessions, per-thread serialization, stateful routing and owned-resource resolution, state-aware Block Kit approvals, durable execution for slow agent work, guaranteed result-or-error delivery, and structured observability.
  133. - [sync-url-navigation](./sync-url-navigation) — syncs URL query params with app navigation, tabs, and filter state so views are bookmarkable and shareable (`view`, `table`, `q`, deep links, `popstate`).
  134. - [app-ux-paradigms](./app-ux-paradigms) — standard web UX defaults: Esc/backdrop/× for modals, ⌘/Ctrl shortcuts, form save states, tables, menus, and help text for discoverable interactions.
  135. - [data-chatbots](./data-chatbots) — copilots over structured data that **propose** mutations (draft → Apply), not direct writes: prompting, validator allowlists, session memory with DRAFT/APPLIED/IGNORED, version-stale UX, and test matrices.
  136. - [reserved-handle-policy](./reserved-handle-policy) — designs and implements two-tier public username protection with hard platform reservations, administrator-reviewed claims, separator-confusable matching, a source-attributed registry of common names and notable identities, and signup/rename/admin test guidance.
  137. ### Cloudflare Product Work
  138. - [workers-best-practices](./workers-best-practices) — reviews Workers runtime, bindings, streaming, concurrency, and static asset routing.
  139. - [wrangler](./wrangler) — resolves exact CLI commands and configuration changes for the selected environment.
  140. - [durable-objects](./durable-objects) — designs per-key coordination, object-owned state, alarms, and WebSockets.
  141. - [agents-sdk](./agents-sdk) — builds stateful apps with Cloudflare's `agents` package.
  142. - [sandbox-sdk](./sandbox-sdk) — uses `@cloudflare/sandbox` for isolated commands, files, processes, and previews.
  143. - [cloudflare-do-turn-based-multiplayer](./cloudflare-do-turn-based-multiplayer) — handles room identity, reconnects, and concurrent turns in Durable Object games.
  144. - [cloudflare-email-service](./cloudflare-email-service) — handles Cloudflare sending, Email Routing, inbound handlers, and deliverability.
  145. - [cloudflare-one](./cloudflare-one) — guides Zero Trust and SASE configuration using current provider documentation.
  146. - [cloudflare-one-migrations](./cloudflare-one-migrations) — plans migration from VPN, SWG, and other SASE deployments.
  147. - [turnstile-spin](./turnstile-spin) — adds or repairs a widget and mandatory server-side verification on a scoped form.
  148. - [diy-netlify](./diy-netlify) — builds isolated pull-request previews with the existing hosting provider.
  149. ### App Design And Interaction
  150. - [design-preferences](./design-preferences) — applies swyx's shared-shell, density, typography, and color preferences.
  151. - [mobile-native](./mobile-native) — covers iPhone/iPad UX, notifications, signing, delivery, and device verification.
  152. - [mobile-webapp-ux](./mobile-webapp-ux) — improves phone layouts and touch flows while preserving desktop usability.
  153. - [website-searchbar](./website-searchbar) — implements universal search, catalog coverage, fuzzy typeahead, and responsiveness.
  154. - [location-input](./location-input) — builds accessible city/country autocomplete.
  155. - [long-running-operation-ux](./long-running-operation-ux) — designs visible progress, cancellation, recovery, and result handoff for slow actions.
  156. - [media-heavy-workflows](./media-heavy-workflows) — designs reference attachment, generation history, and draft-to-publish media workspaces.
  157. - [data-visualization-quality](./data-visualization-quality) — selects analytically faithful charts and tables.
  158. - [web-perf](./web-perf) — investigates a named web performance problem with current browser measurements.
  159. - [cli-ux](./cli-ux) — reviews CLI interaction, output, credentials, and mutation behavior.
  160. ### Agent Execution And Workflow Reuse
  161. - [programmatic-agents](./programmatic-agents) — runs supported coding-agent CLIs with usage, latency, cost, and trace logging; [programmatic-codex](./programmatic-codex) is a compatibility pointer.
  162. - [ai-engineering](./ai-engineering) — repairs unreliable structured outputs, retries, fan-out, caching, and telemetry.
  163. - [live-ai-pipelines](./live-ai-pipelines) — adds partial results, resume, and atomic publication to long AI workflows.
  164. - [babysit-runs](./babysit-runs) — keeps an authorized unattended job progressing through completion.
  165. - [resilient-computer-use](./resilient-computer-use) — observes, acts, and verifies UI state, with isolated tab-bound Chrome control and recovery.
  166. - [coordinate](./coordinate) — coordinates bounded helpers and advisers when delegation is useful.
  167. - [next-steps](./next-steps) — turns open directions into actionable choices.
  168. - [keep-it-simple](./keep-it-simple) — identifies unnecessary complexity and proposes concrete simplifications.
  169. - [future-only](./future-only) — applies explicitly authorized breaking changes without preserving obsolete interfaces.
  170. - [generalize](./generalize) — extracts transferable lessons and broader fixes when requested.
  171. - [provision-model-keys](./provision-model-keys) — provisions or rotates app-specific provider credentials under saved authorization.
  172. ### AI DevRel
  173. AI DevRel is the end-to-end publication bundle: capture technical work, acquire
  174. and transform source media, extract the useful story, publish dense written and
  175. video explanations, operate distribution channels, and measure what changed.
  176. Start with the orchestrator for a multi-stage job, then route each stage to the
  177. narrowest atomic skill.
  178. #### Technical blogging
  179. - [ai-readme](./ai-readme) — turns a repository into a progressive, executable explanation for an explicitly chosen reader, with a verified first result, one stable example, honest tradeoffs, and a context-isolated cold read.
  180. - [ai-devblog](./ai-devblog) — routes technical material into the right story mode and weight, aligns on the reader and belief change, preserves primary evidence, and edits the result for clarity and human interest before verified publication.
  181. - [blog-system-design](./blog-system-design) — designs dense technical blog systems: index and section pages, compact typography, full-text `/` search, resizable article index rails, responsive floating TOCs, reusable explanatory components, and information-dense media policy.
  182. #### Writing And Corpus Publishing
  183. - [tweet-writing](./tweet-writing) — drafts tweets and quote posts using peer examples, source material, and the account's voice.
  184. - [swyx-writing](./swyx-writing) — applies the shared nonfiction voice and three editing passes.
  185. - [research-grounded-writing](./research-grounded-writing) — produces finished nonfiction backed by primary sources.
  186. - [person-profile-writing](./person-profile-writing) — resolves identity and writes source-grounded biographies.
  187. - [video-talk-to-essay](./video-talk-to-essay) — turns a recording and transcript into an illustrated technical article.
  188. - [media-to-wiki-pipeline](./media-to-wiki-pipeline) — ingests source corpora into readers, profiles, and topic pages with resumable subset refreshes.
  189. - [pulp-fiction-writing](./pulp-fiction-writing) — writes and revises serialized genre fiction.
  190. - [latent-space-thumbnail-director](./latent-space-thumbnail-director) — directs Latent Space and FDE thumbnails using saved brand resources.
  191. - [face-matching](./face-matching) — verifies identities and face references across media collections.
  192. #### Media acquisition and transformation
  193. - [media-transform](./media-transform) — orchestrates video pipelines across download, upload, transcription, chapters, thumbnails, and title testing by routing to the right atomic skill for each stage.
  194. - [download-video](./download-video) — downloads embedded videos from web pages by resolving the real player URL and calling `yt-dlp` with the right referer/origin headers.
  195. - [download-x-video](./download-x-video) — downloads X/Twitter post videos with `yt-dlp`, including HLS streams and reliable final-path detection.
  196. - [zoom-download](./zoom-download) — downloads Zoom cloud recordings, verifies filenames/file types, and supports ffmpeg-based content analysis.
  197. #### Transcription, extraction, and repurposing
  198. - [transcribe-anything](./transcribe-anything) — transcribes audio and video files using pluggable ASR backends including local Whisper, whisperX, faster-whisper, OpenAI, Groq, Deepgram, AssemblyAI, Gemini, and Hugging Face models.
  199. - [conference-transcribe](./conference-transcribe) — splits long conference livestreams or YouTube videos into per-talk transcripts using chapter timestamps, segment transcription, and LLM cleanup.
  200. - [multimodal-extraction](./multimodal-extraction) — turns local videos or video URLs into Markdown timelines with slide screenshots, key frames, and transcript spans aligned by timestamp.
  201. - [summarize-anything](./summarize-anything) — recursively summarizes long text with pluggable LLM backends and can emit executive summaries, YouTube descriptions, chapters, posts, titles, thumbnail prompts, blog outlines, and pull quotes.
  202. - [podcast-publishing-assistant](./podcast-publishing-assistant) — turns podcasts, interviews, panels, and long-form audio/video into transcripts, summaries, chapter markers, show notes, titles, descriptions, and promo copy.
  203. #### YouTube operations
  204. - [youtube-channel-operator](./youtube-channel-operator) — designs full-power multi-channel YouTube operators with typed Data/Analytics/Reporting/Live API access, transcript-derived viewer packages, paid-versus-organic analysis, guided Slack approvals, exact-channel OAuth isolation, immutable external-action audit, Studio-only handoffs, and post-change measurement.
  205. - [youtube-api](./youtube-api) — manages YouTube videos programmatically through the YouTube Data API v3, including uploads, thumbnails, metadata updates, and channel video listing.
  206. - [youtube-publish](./youtube-publish) — publishes videos on YouTube, edits titles/descriptions/timestamps, assigns playlists, and manages YouTube Studio metadata workflows.
  207. - [youtube-studio-batch-upload](./youtube-studio-batch-upload) — batches YouTube Studio uploads from Airtable or local video submissions, with source download recovery, metadata staging, unlisted visibility, playlist tagging, save verification, and blocked-row reporting.
  208. - [youtube-studio-computer-use](./youtube-studio-computer-use) — automates live YouTube Studio cleanup through Chrome/Computer Use: thumbnails, schedules, playlist fixes, visibility, save-state recovery, and DOM-assisted edit pages.
  209. - [youtube-thumbnails](./youtube-thumbnails) — creates AI-generated YouTube thumbnails with prompt engineering, image generation, compression, and upload guidance.
  210. - [thumbnail-extraction](./thumbnail-extraction) — extracts interesting video frames, face crops, presentation slides, and transparent cutouts for thumbnail compositing.
  211. ### Web And Social Scraping
  212. - [twitter-x-scraping](./twitter-x-scraping) — scrapes public Twitter/X profile and list timelines through Nitter-compatible mirror HTML, with cursor pagination, raw JSON persistence, and anti-bot fallback guidance. Does not treat public mirrors as a reliable source for a user's following graph.
  213. ### Workspace And Organizer Administration
  214. - [gsuite-setup](./gsuite-setup) — configures Google Workspace sharing, Groups, delegation, and collaboration settings.
  215. - [sessionize-automation](./sessionize-automation) — inspects and verifies authenticated Sessionize organizer changes.
  216. ### Conference And Event Operations
  217. - [accelevents-api](./accelevents-api) — reads and updates AI Engineer Europe speaker records through the Accelevents REST API while preserving full speaker payloads.
  218. - [accelevents-speaker-sync](./accelevents-speaker-sync) — syncs website speaker, session, schedule, room, track, and headshot changes back to Accelevents for AI Engineer Europe.
  219. - [conference-developer-endpoints](./conference-developer-endpoints) — adds and reviews developer-facing conference endpoints such as `llms.txt`, `sessions.json`, `speakers.json`, and MCP routes.
  220. - [europe-developer-api](./europe-developer-api) — works with AI Engineer Europe developer endpoints, public schedule JSON, speakers JSON, MCP access, and the local `aieng` CLI.
  221. - [schedule-design](./schedule-design) — builds polished conference schedule views with React grids, filters, modals, favorites, sticky layouts, and normalized data.
  222. - [sync-accelevents](./sync-accelevents) — pulls Accelevents speaker headshots, social data, bios, and schedule metadata into local conference source data.
  223. - [testing-schedule-preview](./testing-schedule-preview) — tests the AI Engineer Europe internal Bun schedule preview and public schedule page workflows.
  224. - [web-animation-perf](./web-animation-perf) — debugs jank, layout thrash, and drift in JS-driven CSS animation across AI Engineer conference sites.
  225. ## Repo Shape
  226. - One canonical skill per top-level folder; `programmatic-codex` is a compatibility directory.
  227. - Every canonical skill includes `SKILL.md` with YAML `name` and `description` metadata.
  228. - Optional `references/`, `scripts/`, `assets/`, and `agents/openai.yaml` carry detailed guidance, executables, resources, or agent UI metadata.
  229. - Add scripts only when they make the workflow more reliable or repeatable.
  230. - Keep auxiliary docs minimal; the skill body should carry the agent-facing workflow.
  231. Click into each folder for the detailed workflow, prerequisites, and command examples.
  232. The index above covers the tracked canonical skills. HyperFrames and its
  233. companion media workflows have been removed from this collection; the remaining
  234. skills are stored in this repository.
  235. ## Recent Workflow Changes
  236. Recent commits added [review-thread](./review-thread) progress assessments and
  237. [mobile-native](./mobile-native) preferences and device guidance. Design
  238. exploration now uses distinct Mobbin-informed briefs and honors delegated
  239. choices without repeated approval. Chrome workflows favor isolated tab-bound
  240. control. Shared execution guidance is more agent-neutral, while writing and
  241. wiki workflows preserve source evidence and support incremental corpus updates.
  242. See `git log --oneline -8` for the current commit history.
  243. ## Skill acceptance model
  244. A skill is an advisory lens, not an implicit acceptance gate. Author and review
  245. skills with this model:
  246. ```text
  247. User outcome
  248. + higher-level invariants
  249. + risks created by this action
  250. = blocking acceptance criteria
  251. Everything else is advice or follow-up.
  252. ```
  253. Classify meaningful instructions by their actual force:
  254. | Class | Meaning |
  255. | --- | --- |
  256. | **Invariant** | Must never be violated, such as authorization, privacy, secret handling, destructive-target clarity, or user-data integrity. |
  257. | **Action-required** | Intrinsic to the named task type; without it the requested result is not correct or usable. |
  258. | **Risk-triggered gate** | Blocking only when the skill names a concrete risk introduced by the proposed action. |
  259. | **Recommendation** | A useful default that may be skipped without blocking completion. |
  260. | **Opportunity** | An adjacent improvement or follow-up outside the current critical path. |
  261. Write the class directly when prose could otherwise make a recommendation sound
  262. mandatory. A comprehensive checklist is not a demand to satisfy every item:
  263. tell agents to select only relevant items and state the concrete risk before
  264. promoting one to a gate.
  265. For bounded tasks, a skill should normally add no more than one or two blocking
  266. criteria. Exceed that budget only for a direct correctness, privacy, security,
  267. data-integrity, or irreversible-action risk. Match evidence to impact: a local
  268. documentation edit may need formatting and link checks; a CLI contract fix
  269. needs focused process or contract tests; one Worker repair needs component
  270. checks, health, one bounded reproduction, and rollback evidence; a schema or
  271. data mutation warrants stronger integrity and repair proof; a destructive or
  272. externally consequential action requires exact targets and explicit authority.
  273. Do not silently expand scope. Adjacent improvements, checklist findings, and
  274. residual issues are observations or follow-ups unless necessary to make the
  275. requested result correct, safe, or usable. Define a stop condition for workflows
  276. that can otherwise accumulate proof or retries, and allow completion when that
  277. condition is met.
  278. Do not make a broken control plane approve or execute its own repair when a
  279. documented lower-level operator path exists. Preserve that path's authorization,
  280. target-resolution, rollback, and evidence requirements. Coordinate only for
  281. concrete overlap: the same files with likely merge conflicts, production
  282. resource, migration sequence, or externally mutable object. Read-only work and
  283. unrelated components do not need a global mutex or continual peer updates.
  284. Users may simplify recommendations and authorize documented break-glass paths.
  285. They cannot waive higher-level constraints around secrets, destructive
  286. ambiguity, unauthorized external action, privacy, or irreversible user-data
  287. loss.
  288. ## Validating Skills
  289. The repository pins PyYAML in `pyproject.toml` and `uv.lock`. With
  290. [uv](https://docs.astral.sh/uv/) installed, prepare those dependencies from the
  291. repository root:
  292. ```bash
  293. uv sync --locked
  294. ```
  295. This prepares `.venv` without changing the lockfile. Validate a changed skill
  296. with the repository's shared [Agent Skills](https://agentskills.io/specification)
  297. metadata validator:
  298. ```bash
  299. uv run --locked python scripts/validate_skill.py review-thread
  300. ```
  301. A successful run prints `review-thread: Skill is valid!`. The validator checks
  302. standard metadata, including `compatibility`, and requires custom fields such as
  303. `version` inside `metadata`. Keep shared frontmatter portable; loader-specific
  304. invocation hints can be documented in the skill body. Older environment-supplied
  305. validators may reject valid standard fields.
  306. Validation checks metadata and structure, not workflow correctness. For changed
  307. skills, also check relative links, read referenced files, and run the focused
  308. script tests named in that skill when applicable.
  309. Before contributing, read [AGENTS.md](./AGENTS.md), preserve unrelated edits,
  310. and use `git diff --check` for whitespace errors. Keep top-level discovery,
  311. working reference paths, focused trigger descriptions, and the acceptance model
  312. above intact. Review substantial behavior concerns explicitly rather than
  313. silently expanding a documentation change into implementation.
new-mac-setup/scripts/06-dotfiles.sh · Bundled script · 576 lines · 2 findings marked
  1. #!/bin/bash
  2. # =============================================================================
  3. # SWYX 2025 NEW MAC SETUP — STEP 6: Dotfiles (.zshrc + configs)
  4. # =============================================================================
  5. # Dotfiles gist: https://gist.github.com/swyxio/7fa1009e460ecb818d5e6d9ca4616a05
  6. # =============================================================================
  7. set -e
  8. echo "📄 SWYX NEW MAC SETUP — Step 6: Dotfiles"
  9. echo "=========================================="
  10. echo ""
  11. # =============================================================================
  12. # BACKUP EXISTING .zshrc
  13. # =============================================================================
  14. if [ -f "$HOME/.zshrc" ]; then
  15. echo "📋 Backing up existing .zshrc to .zshrc.backup..."
  16. cp "$HOME/.zshrc" "$HOME/.zshrc.backup"
  17. fi
  18. # =============================================================================
  19. # WRITE .zshrc
  20. # =============================================================================
  21. echo "📝 Writing .zshrc..."
  22. cat > "$HOME/.zshrc" << 'ZSHRC'
  23. # =============================================================================
  24. # SWYX .zshrc — 2025 edition
  25. # Merged from gist + 2025 improvements (uv, colima, powerlevel10k, z)
  26. # Dotfiles gist: https://gist.github.com/swyxio/7fa1009e460ecb818d5e6d9ca4616a05
  27. # =============================================================================
  28. # --- Oh-My-ZSH ---
  29. export ZSH="$HOME/.oh-my-zsh"
  30. ZSH_THEME="powerlevel10k/powerlevel10k"
  31. # To switch back to original robbyrussell theme: ZSH_THEME="robbyrussell"
  32. plugins=(
  33. git
  34. zsh-autosuggestions
  35. zsh-syntax-highlighting
  36. z
  37. )
  38. source $ZSH/oh-my-zsh.sh
  39. # --- Homebrew ---
  40. eval "$(/opt/homebrew/bin/brew shellenv)"
  41. # --- fnm (Fast Node Manager) - replaces nvm ---
  42. eval "$(fnm env --use-on-cd)"
  43. # --- uv (Python) - Astral's replacement for pip/pyenv/conda ---
  44. [ -f "$HOME/.local/bin/env" ] && source "$HOME/.local/bin/env"
  45. # --- z (directory jumping) ---
  46. [ -f /opt/homebrew/etc/profile.d/z.sh ] && . /opt/homebrew/etc/profile.d/z.sh
  47. # =============================================================================
  48. # ALIASES — File Management
  49. # =============================================================================
  50. alias mv="mv -iv"
  51. alias cp="cp -iv"
  52. alias lsd="ls -ltr"
  53. # =============================================================================
  54. # ALIASES — Shell Management
  55. # =============================================================================
  56. alias zs="source ~/.zshrc"
  57. alias zshrc="vi ~/.zshrc"
  58. # =============================================================================
  59. # ALIASES — Git
  60. # =============================================================================
  61. alias gclone="git clone --depth 1 "
  62. alias gpom="git push origin main"
  63. alias gpo="git push origin "
  64. alias gset="git remote set-url"
  65. alias gs="git status"
  66. alias gadmit="git add . && git commit -m"
  67. alias gc="git checkout"
  68. alias gcb="git checkout -b "
  69. alias gupdatefork="git fetch upstream && git checkout master && git rebase upstream/master"
  70. alias gcm="git checkout main 2> /dev/null || git checkout master"
  71. # git log with fancy formatting
  72. alias gl="git log --graph --abbrev-commit --decorate --format=format:'%C(bold blue)%h%C(reset) - %C(bold green)(%ar)%C(reset) %C(white)%s%C(reset) %C(dim white)- %an%C(reset)%C(bold yellow)%d%C(reset)' --all"
  73. alias glr='gl -n 7'
  74. # git branch cleanup
  75. alias git-clean-merged="git branch --merged master | grep -v 'master$' | xargs git branch -d"
  76. alias git-clean-remote-merged="git branch -r --merged | grep -v master | sed 's/origin\///' | xargs -n 1 git push --delete origin"
  77. # git commit summary
  78. alias gitsum='git log --pretty=format:"* %s" --author `git config user.email`'
  79. # =============================================================================
  80. # ALIASES — npm/pnpm
  81. # =============================================================================
  82. alias ni="npm install"
  83. alias nr="npm run"
  84. alias pi="pnpm install"
  85. alias pr="pnpm run"
  86. alias ncu="npm-check-updates"
  87. alias run="npm run"
  88. alias yb="pnpm run build || npm run build"
  89. alias ys="pnpm run start || npm run start"
  90. # =============================================================================
  91. # ALIASES — Editors & Tools
  92. # =============================================================================
  93. alias cc="claude --dangerously-skip-permissions"
  94. alias co="codex --dangerously-bypass-approvals-and-sandbox"
  95. alias fucking=sudo
  96. alias ytdlp='yt-dlp -f "bestvideo[ext=mp4][height<=720]+bestaudio[ext=m4a]/best[ext=mp4]/best" '
  97. alias json_pretty="pbpaste | python -m json.tool | pbcopy"
  98. # =============================================================================
  99. # ALIASES — Docker via Colima
  100. # =============================================================================
  101. alias dstart="colima start"
  102. alias dstop="colima stop"
  103. alias dc="docker-compose"
  104. # =============================================================================
  105. # FUNCTIONS — Directory
  106. # =============================================================================
  107. # mkcdir: create directory and cd into it
  108. function mkcdir () {
  109. mkdir -p -- "$1" &&
  110. cd -P -- "$1"
  111. }
  112. # tree: show directory tree with sane defaults
  113. function t() {
  114. tree -I '.git|node_modules|bower_components|.DS_Store' --dirsfirst --filelimit 15 -L ${1:-3} -aC $2
  115. }
  116. # =============================================================================
  117. # FUNCTIONS — AI Commit Message Generation (ggadmit)
  118. # =============================================================================
  119. unalias ggadmit 2>/dev/null
  120. ggadmit() {
  121. git add .
  122. local tmpfile=$(mktemp)
  123. printf 'Generate a concise git commit message for these changes...\n%s' "$(git diff --cached)" > "$tmpfile"
  124. local msg=$(devin --model claude-sonnet-4.5 -p --prompt-file "$tmpfile")
  125. rm -f "$tmpfile"
  126. if [ -z "$msg" ]; then
  127. echo "Error: devin returned empty commit message" >&2
  128. return 1
  129. fi
  130. git commit -m "$msg"
  131. }
  132. # =============================================================================
  133. # FUNCTIONS — AI Command Generation (please)
  134. # =============================================================================
  135. needs() {
  136. which "$1" > /dev/null 2>&1 || { echo "I require $1 but it's not installed. Aborting." >&2; return 1; }
  137. }
  138. platform() {
  139. case "$OSTYPE" in
  140. darwin*) echo "macOS" ;;
  141. linux*) echo "Linux" ;;
  142. *) echo "Unknown" ;;
  143. esac
  144. }
  145. _generate_curl_api_request_for_please() {
  146. local prompt="$1"
  147. local model="${2:-gpt-4}"
  148. local api_key="${OPENAI_API_KEY}"
  149. if [ -z "$api_key" ]; then
  150. echo "Error: OPENAI_API_KEY not set" >&2
  151. return 1
  152. fi
  153. curl -s https://api.openai.com/v1/chat/completions \
  154. -H "Content-Type: application/json" \
  155. -H "Authorization: Bearer $api_key" \
  156. -d "{
  157. \"model\": \"$model\",
  158. \"messages\": [
  159. {\"role\": \"system\", \"content\": \"You are a helpful bash expert. Generate a single bash command to accomplish the task. Output ONLY the command, no explanations.\"},
  160. {\"role\": \"user\", \"content\": \"$prompt\"}
  161. ],
  162. \"temperature\": 0
  163. }" | grep -o '"content":"[^"]*' | head -1 | sed 's/"content":"//'
  164. }
  165. please() {
  166. needs gum
  167. local prompt="$1"
  168. local model="${2:-gpt-4}"
  169. if [ -z "$prompt" ]; then
  170. echo "Usage: please <description> [model]" >&2
  171. return 1
  172. fi
  173. local cmd=$(_generate_curl_api_request_for_please "$prompt" "$model")
  174. if [ -z "$cmd" ]; then
  175. echo "Error: Failed to generate command" >&2
  176. return 1
  177. fi
  178. echo "Generated command:"
  179. echo " $cmd"
  180. echo ""
  181. if gum confirm "Execute this command?"; then
  182. eval "$cmd"
  183. else
  184. echo "Command not executed."
  185. fi
  186. }
  187. # =============================================================================
  188. # PATH & ENVIRONMENT
  189. # =============================================================================
  190. export PATH="$HOME/.local/bin:$PATH"
  191. export PATH="/opt/homebrew/bin:$PATH"
  192. export PATH="/opt/homebrew/share/google-cloud-sdk/bin:$PATH"
  193. # Editor preferences
  194. export EDITOR="cursor"
  195. export VISUAL="cursor"
  196. # =============================================================================
  197. # KEYBOARD BINDINGS
  198. # =============================================================================
  199. bindkey '^\[\[H' beginning-of-line # Home key
  200. bindkey '^\[\[F' end-of-line # End key
  201. bindkey "^\[\[3-" delete-char # Delete key
  202. bindkey '^\[\[1;5D' backward-word # Ctrl+Left
  203. bindkey '^\[\[1;5C' forward-word # Ctrl+Right
  204. # =============================================================================
  205. # POWERLEVEL10K THEME
  206. # =============================================================================
  207. # To customize prompt, run `p10k configure` or edit ~/.p10k.zsh.
  208. [[ ! -f ~/.p10k.zsh ]] || source ~/.p10k.zsh
  209. ZSHRC
  210. echo " ✅ .zshrc written"
  211. # =============================================================================
  212. # WRITE .gitignore_global
  213. # =============================================================================
  214. echo ""
  215. echo "📝 Writing .gitignore_global..."
  216. cat > "$HOME/.gitignore_global" << 'GITIGNORE_GLOBAL'
  217. # =============================================================================
  218. # Global .gitignore for swyx — applied to all repos
  219. # Install: git config --global core.excludesfile ~/.gitignore_global
  220. # =============================================================================
  221. # OS
  222. .DS_Store
  223. .DS_Store?
  224. ._*
  225. .Spotlight-V100
  226. .Trashes
  227. ehthumbs.db
  228. Thumbs.db
  229. # Editors & IDEs
  230. .vscode/
  231. .idea/
  232. *.swp
  233. *.swo
  234. *~
  235. .vim/
  236. .nvim/
  237. .emacs.d/
  238. *.sublime-project
  239. *.sublime-workspace
  240. # Package managers
  241. node_modules/
  242. .pnpm-store/
  243. .yarn/
  244. dist/
  245. build/
  246. .next/
  247. .nuxt/
  248. # Python
  249. __pycache__/
  250. *.py[cod]
  251. *$py.class
  252. *.so
  253. .Python
  254. env/
  255. venv/
  256. ENV/
  257. .venv
  258. pip-log.txt
  259. pip-delete-this-directory.txt
  260. # Ruby
  261. Gemfile.lock
  262. .ruby-version
  263. # Environment
  264. .env
  265. .env.local
  266. .env.*.local
  267. # Logs
  268. *.log
  269. npm-debug.log*
  270. yarn-debug.log*
  271. yarn-error.log*
  272. # Temporary files
  273. *.tmp
  274. *.bak
  275. .cache/
  276. .temp/
  277. # macOS specifics
  278. .AppleDouble
  279. .LSOverride
  280. # IDE settings (keep .vscode/settings.json, but ignore workspace state)
  281. .vscode/settings.json.bak
  282. # Testing
  283. .nyc_output/
  284. coverage/
  285. # Misc
  286. .turbo/
  287. .eslintcache
  288. .stylelintcache
  289. GITIGNORE_GLOBAL
  290. # Install globally
  291. git config --global core.excludesfile ~/.gitignore_global
  292. echo " ✅ .gitignore_global written and configured"
  293. # =============================================================================
  294. # WRITE .vimrc
  295. # =============================================================================
  296. echo ""
  297. echo "📝 Writing .vimrc..."
  298. cat > "$HOME/.vimrc" << 'VIMRC'
  299. " =============================================================================
  300. " SWYX .vimrc — 2025 edition
  301. " =============================================================================
  302. set nocompatible
  303. filetype plugin indent on
  304. syntax on
  305. " Indentation
  306. set expandtab
  307. set tabstop=4
  308. set shiftwidth=4
  309. set softtabstop=4
  310. " Filetype-specific indentation
  311. autocmd FileType json setlocal tabstop=2 shiftwidth=2 softtabstop=2
  312. autocmd FileType html setlocal tabstop=2 shiftwidth=2 softtabstop=2
  313. autocmd FileType css setlocal tabstop=2 shiftwidth=2 softtabstop=2
  314. autocmd FileType javascript setlocal tabstop=2 shiftwidth=2 softtabstop=2
  315. autocmd FileType typescript setlocal tabstop=2 shiftwidth=2 softtabstop=2
  316. autocmd FileType vue setlocal tabstop=2 shiftwidth=2 softtabstop=2
  317. autocmd FileType yaml setlocal tabstop=2 shiftwidth=2 softtabstop=2
  318. autocmd FileType svg setlocal tabstop=2 shiftwidth=2 softtabstop=2
  319. " No backups/swapfiles
  320. set nobackup
  321. set nowritebackup
  322. set noswapfile
  323. " Search
  324. set ignorecase
  325. set smartcase
  326. set incsearch
  327. set hlsearch
  328. " Display
  329. set number
  330. set relativenumber
  331. set cursorline
  332. set wrap
  333. set linebreak
  334. set showmatch
  335. set lazyredraw
  336. " Behavior
  337. set backspace=indent,eol,start
  338. set mouse=a
  339. set ttimeoutlen=10
  340. " Status line
  341. set laststatus=2
  342. set statusline=%F%m%r%h%w\ [%{&ff}]\ %l:%c
  343. " Keybindings: preserve whitespace in insert mode
  344. inoremap <C-U> <C-G>u<C-U>
  345. inoremap <C-W> <C-G>u<C-W>
  346. " Quick save
  347. nnoremap <leader>w :w<CR>
  348. nnoremap <leader>q :q<CR>
  349. " Navigation
  350. nnoremap j gj
  351. nnoremap k gk
  352. " FZF integration (if available)
  353. if has('nvim') || has('gui_running')
  354. set rtp+=/opt/homebrew/opt/fzf
  355. endif
  356. VIMRC
  357. echo " ✅ .vimrc written"
  358. # =============================================================================
  359. # WRITE VSCode/Cursor settings
  360. # =============================================================================
  361. echo ""
  362. echo "📝 Writing VSCode/Cursor settings..."
  363. # Create VSCode user settings directory
  364. mkdir -p "$HOME/Library/Application Support/Code/User"
  365. cat > "$HOME/Library/Application Support/Code/User/settings.json" << 'VSCODE_SETTINGS'
  366. {
  367. "editor.fontFamily": "Meslo LG M for Powerline",
  368. "editor.fontSize": 14,
  369. "editor.formatOnSave": true,
  370. "editor.codeActionsOnSave": {
  371. "source.fixAll.eslint": "explicit",
  372. "source.fixAll.prettier": "explicit"
  373. },
  374. "editor.defaultFormatter": "esbenp.prettier-vscode",
  375. "editor.insertSpaces": true,
  376. "editor.tabSize": 2,
  377. "files.trimTrailingWhitespace": true,
  378. "files.insertFinalNewline": true,
  379. "files.trimFinalNewlines": true,
  380. "files.exclude": {
  381. "**/.git": true,
  382. "**/.DS_Store": true,
  383. "**/.next": true,
  384. "**/node_modules": true,
  385. "**/.venv": true
  386. },
  387. "search.exclude": {
  388. "**/.git": true,
  389. "**/.next": true,
  390. "**/node_modules": true,
  391. "**/dist": true,
  392. "**/.venv": true
  393. },
  394. "workbench.colorTheme": "GitHub Dark",
  395. "workbench.iconTheme": "GitHub Modern",
  396. "terminal.integrated.fontFamily": "Meslo LG M for Powerline",
  397. "terminal.integrated.fontSize": 14,
  398. "terminal.integrated.defaultProfile.osx": "zsh",
  399. "terminal.integrated.profiles.osx": {
  400. "zsh": {
  401. "path": "/bin/zsh",
  402. "args": ["-l"]
  403. }
  404. },
  405. "git.ignoreLimitWarning": true,
  406. "[json]": {
  407. "editor.tabSize": 2
  408. },
  409. "[yaml]": {
  410. "editor.tabSize": 2
  411. },
  412. "[markdown]": {
  413. "editor.wordWrap": "on",
  414. "editor.formatOnSave": false
  415. },
  416. "extensions.ignoreRecommendations": false,
  417. "telemetry.telemetryLevel": "off"
  418. }
  419. VSCODE_SETTINGS
  420. echo " ✅ VSCode/Cursor settings written to ~/Library/Application Support/Code/User/settings.json"
  421. # Create Cursor-specific settings directory (if using Cursor)
  422. if [ -d "$HOME/Library/Application Support/Cursor" ]; then
  423. mkdir -p "$HOME/Library/Application Support/Cursor/User"
  424. cp "$HOME/Library/Application Support/Code/User/settings.json" "$HOME/Library/Application Support/Cursor/User/settings.json"
  425. echo " ✅ Also copied to Cursor settings directory"
  426. fi
  427. # =============================================================================
  428. # GHOSTTY CONFIG
  429. # =============================================================================
  430. echo ""
  431. echo "📝 Writing Ghostty config..."
  432. GHOSTTY_DIR="$HOME/.config/ghostty"
  433. mkdir -p "$GHOSTTY_DIR"
  434. cat > "$GHOSTTY_DIR/config" << 'GHOSTTYCONF'
  435. # =============================================================================
  436. # Ghostty config — swyx 2025
  437. # cmux also reads this file for themes/fonts/colors
  438. # =============================================================================
  439. # Font
  440. font-family = "Meslo LG M for Powerline"
  441. font-size = 14
  442. # Theme
  443. theme = dark:GruvboxDark,light:GruvboxLight
  444. # Window
  445. window-padding-x = 8
  446. window-padding-y = 4
  447. window-decoration = true
  448. macos-titlebar-style = tabs
  449. # Shell
  450. shell-integration = zsh
  451. # Cursor
  452. cursor-style = block
  453. cursor-style-blink = false
  454. # Scrollback
  455. scrollback-limit = 10000
  456. # Clipboard
  457. clipboard-read = allow
  458. clipboard-write = allow
  459. # Mouse
  460. mouse-hide-while-typing = true
  461. # Quick terminal (drop-down)
  462. keybind = global:cmd+grave_accent=toggle_quick_terminal
  463. GHOSTTYCONF
  464. echo " ✅ Ghostty config written to ~/.config/ghostty/config"
  465. echo " 💡 cmux will also read this config for themes, fonts, and colors"
  466. # =============================================================================
  467. # COMPLETE
  468. # =============================================================================
  469. echo ""
  470. echo "✅ Step 6 complete! All dotfiles written:"
  471. echo " • ~/.zshrc — Full shell config with all aliases & functions"
  472. echo " • ~/.gitignore_global — Global git ignore patterns"
  473. echo " • ~/.vimrc — Vim configuration"
  474. echo " • ~/Library/Application Support/Code/User/settings.json — VSCode/Cursor"
  475. echo " • ~/.config/ghostty/config — Ghostty terminal emulator"
  476. echo ""
  477. echo "Next steps:"
  478. echo " • Run: source ~/.zshrc (or open a new terminal)"
  479. echo " • If using Cursor/VSCode, reload the window (Cmd+K, Cmd+R)"
  480. echo " • Run ./07-macos-settings.sh next."

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.