Skip to content
Back to search

builder-skills

MarketplaceWorks with Claude CodeGitHub

Public Builder skills for docs-first discipline, autonomous execution, cross-agent review, local-first Rewind memory, efficient agent workflows, clear status recaps, Agent-Native workflow apps, and experimental configurable software factory workflows.

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/builderio/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/builderio/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/builderio-skills/badge.svg)](/listings/builderio-skills)

HTML

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

Install

Install instructions from the README

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

npx @agent-native/skills@latest add

See the [full CLI docs below](#install).

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.

.claude-plugin/marketplace.json · Manifest · 24 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. {
  2. "name": "builder-skills",
  3. "owner": {
  4. "name": "Builder.io",
  5. "url": "https://www.builder.io"
  6. },
  7. "description": "Public Builder skills for docs-first discipline, autonomous execution, cross-agent review, local-first Rewind memory, efficient agent workflows, clear status recaps, Agent-Native workflow apps, and experimental configurable software factory workflows.",
  8. "plugins": [
  9. {
  10. "name": "builder-skills",
  11. "source": "./",
  12. "displayName": "Builder Skills",
  13. "description": "Public Builder skills for docs-first discipline, autonomous execution, cross-agent review, local-first Rewind memory, efficient agent workflows, clear status recaps, Agent-Native workflow apps, and experimental configurable software factory workflows.",
  14. "author": {
  15. "name": "Builder.io",
  16. "url": "https://www.builder.io"
  17. },
  18. "homepage": "https://github.com/BuilderIO/skills",
  19. "repository": "https://github.com/BuilderIO/skills",
  20. "license": "MIT",
  21. "keywords": ["builder", "skills", "agent-workflows", "subagents"]
  22. }
  23. ]
  24. }
README.md · README · 534 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. # Skills for agents
  2. Small, composable skills for your favorite agent.
  3. ### Quick install recommended skills
  4. ```sh
  5. npx @agent-native/skills@latest add
  6. ```
  7. See the [full CLI docs below](#install).
  8. ## Skills At A Glance
  9. - [`/factory`](#factory) (experimental) - Configure connected
  10. sources, prompts, schedules, and delivery policies.
  11. - [`/factory-collect`](#factory-collect) - Collect
  12. feedback, product telemetry, errors, and issues.
  13. - [`/factory-babysit-pr`](#factory-babysit-pr) -
  14. Follow one authorized PR through checks and review.
  15. - [`/factory-review-prs`](#factory-review-prs) -
  16. Review a PR queue with separate reply, approval, and merge criteria.
  17. - [`/factory-ship`](#factory-ship) - Publish
  18. delivery work under configured verification and merge policies.
  19. - [`/factory-lookback`](#factory-lookback) - Find
  20. recurring signals and systemic fixes across a chosen time window.
  21. - [`/factory-human-digest`](#factory-human-digest) (experimental) - Summarize
  22. work that still needs human judgment across configured sources.
  23. - [`/factory-watchdog`](#factory-watchdog) - Find
  24. stalled delivery work and notify when a concrete next step is due.
  25. - [`/factory-recover`](#factory-recover) - Resume
  26. interrupted work when its authorization and worktree are still valid.
  27. - [`/an`](#an) - Open and operate Agent-Native apps beside the conversation.
  28. - [`/webmcp`](#webmcp) - Open web apps in the built-in browser and use MCP tools first.
  29. - [`/visual-plan`](#visual-plan) - Turn text plans into rich visual plans.
  30. - [`/visual-recap`](#visual-recap) - Turn diffs into interactive visual recaps.
  31. - [`/visual-edit`](#visual-edit) - Open a running local app for visual editing.
  32. - [`/remix`](#remix) - Rebuild part of another website in yours, with live knobs.
  33. - [`/rewind`](#rewind) - Recover recent local screen context through Clips Desktop.
  34. - [`/agent-watchdog`](#agent-watchdog) - Audit another agent's work.
  35. - [`/plan-arbiter`](#plan-arbiter) - Compare competing plans and choose a direction.
  36. - [`/plow-ahead`](#plow-ahead) - Keep working through ordinary ambiguity.
  37. - [`/efficient-orchestrator`](#efficient-orchestrator) - Keep an expensive model on judgment while cheap agents do the heavy lifting (aliases: `/efficient-frontier`, `/efficient-fable`).
  38. - [`/stay-within-limits`](#stay-within-limits) - Track usage limits before long-running work.
  39. - [`/quick-recap`](#quick-recap) - End work with a clear status signal.
  40. - [`/read-the-damn-docs`](#read-the-damn-docs) - Check authoritative docs before guessing.
  41. - [`/turn-into-app`](#turn-into-app) - Turn a thread, skill, spreadsheet, or Claude/ChatGPT project into a visual Agent-Native app.
  42. ## Skill Details
  43. ### /factory
  44. **Experimental.** Start with the [Factory guide](docs/factory/README.md), then
  45. open the [full Factory skill](skills/factory/SKILL.md) for agent instructions.
  46. Factory is an experimental set of agent skills for turning feedback, product
  47. telemetry, errors, and delivery signals into policy-gated software changes.
  48. The modules can collect current reports, look back for recurring failures, and
  49. review or ship changes. Configure schedules and separate rules for fixing,
  50. replying, reviewing, approving, merging, deploying, and closing.
  51. ```mermaid
  52. flowchart LR
  53. sources["Feedback and telemetry"] --> collect["/factory-collect"]
  54. sources --> lookback["/factory-lookback<br/>recurring patterns"]
  55. collect --> agent["Factory workflow"]
  56. lookback --> agent
  57. agent -->|"policy allows"| fix["Isolated fix<br/>checks and verification"]
  58. agent -->|"unclear or out of scope"| human["Human decision"]
  59. fix --> pr["Pull request"]
  60. pr -->|"approval and merge rules"| ship["Ship"]
  61. pr -->|"needs judgment"| human
  62. ship -->|"if allowed"| source["Reply or close source item"]
  63. ```
  64. #### Configuration at a glance
  65. This abbreviated example uses placeholder scopes and connected tools. Factory
  66. does not install these integrations or create the schedule for you.
  67. ```yaml
  68. version: 1
  69. timezone: UTC
  70. sources:
  71. - id: support
  72. provider: slack
  73. scope: channel-id
  74. - id: errors
  75. provider: sentry
  76. scope: organization/project
  77. repositories:
  78. - id: app
  79. provider: github
  80. remote: example/project
  81. workflows:
  82. collect:
  83. enabled: true
  84. sources: [support, errors]
  85. reply:
  86. mode: criteria
  87. require: [one missing detail blocks triage or verification]
  88. tone: warm and direct
  89. guidance: Ask one targeted question, then re-triage when an answer arrives.
  90. lookback:
  91. enabled: true
  92. schedule: monthly
  93. window: last 30 days
  94. sources: [support, errors]
  95. implement:
  96. mode: manual
  97. human-digest:
  98. enabled: true
  99. schedule: weekly
  100. window: last 7 days
  101. repositories: [app]
  102. sources: [support, errors]
  103. include: [pull-requests, issues, feedback, errors]
  104. granularity: balanced
  105. skill_prompts:
  106. factory-ship: |
  107. Keep release summaries concise and link the verified change.
  108. factory-human-digest: |
  109. Group repeated UX concerns, but include a link to every source item.
  110. ```
  111. Each Factory skill reads its own optional prompt entry as additional project
  112. guidance. Prompt text cannot authorize an action that its separate policy
  113. blocks.
  114. The [configuration reference](docs/factory/configuration.md) explains the
  115. available YAML fields, source patterns, policies, and host limitations.
  116. ### /factory-collect
  117. Start with the [Factory guide](docs/factory/README.md), then read the
  118. [full collect skill](skills/factory-collect/SKILL.md).
  119. ### /factory-babysit-pr
  120. Start with the [Factory guide](docs/factory/README.md), then read the
  121. [full PR babysitting skill](skills/factory-babysit-pr/SKILL.md).
  122. ### /factory-review-prs
  123. Start with the [Factory guide](docs/factory/README.md), then read the
  124. [full PR review skill](skills/factory-review-prs/SKILL.md).
  125. ### /factory-ship
  126. Start with the [Factory guide](docs/factory/README.md), then read the
  127. [full shipping skill](skills/factory-ship/SKILL.md).
  128. ### /factory-lookback
  129. Start with the [Factory guide](docs/factory/README.md), then read the
  130. [full lookback skill](skills/factory-lookback/SKILL.md).
  131. ### /factory-human-digest
  132. **Experimental.** Start with the [Factory guide](docs/factory/README.md), then
  133. read the [full human digest skill](skills/factory-human-digest/SKILL.md). It
  134. defaults to all configured human-pending work from the last 7 days at balanced
  135. detail; ask for a narrower source type, another date range, or brief or detailed
  136. output.
  137. ### /factory-watchdog
  138. Start with the [Factory guide](docs/factory/README.md), then read the
  139. [full watchdog skill](skills/factory-watchdog/SKILL.md).
  140. ### /factory-recover
  141. Start with the [Factory guide](docs/factory/README.md), then read the
  142. [full recovery skill](skills/factory-recover/SKILL.md).
  143. ### [`/an`](skills/an/SKILL.md)
  144. Open a granted Agent-Native app in the host's inline MCP App or built-in browser
  145. surface, then keep the agent connected to the app's live UI and screen state.
  146. For example, run `/an slides` to open Slides, sign in in that browser
  147. surface, and then ask the agent to create or update the deck. Focused requests
  148. such as `make this bigger` use the Slides selection state and a bounded
  149. readback-verified edit.
  150. The plugin registers Dispatch MCP automatically for Claude and Codex at
  151. `https://dispatch.agent-native.com/mcp`. In ChatGPT, add that same URL as
  152. an OAuth custom connector. The host may need a reload or connector rescan after
  153. installation.
  154. ### [`/webmcp`](skills/webmcp/README.md)
  155. Open a requested URL in the host's built-in browser and use the page's MCP or
  156. WebMCP tools before browser UI automation for app communication or edits. For
  157. example, run `/webmcp slides.agent-native.com make me a new deck about onboarding`
  158. to open Slides and complete the request in that tab. The skill waits for the
  159. user to sign in when needed, then calls the page's `window.__agentNativeWebMcp`
  160. helper from the host's JavaScript evaluator (or a host WebMCP bridge when one
  161. is exposed): one screen read, one smallest mutation, one readback. If neither
  162. is available, it stops before a state-changing click or type fallback and
  163. reports the limitation.
  164. ### [`/visual-edit`](skills/visual-edit/README.md)
  165. Open a running local app in Design as URL-backed iframe screens for visual
  166. inspection, route-state exploration, and source-backed editing.
  167. Unlike a static mockup, the canvas stays connected to the app's real routes and
  168. the local bridge. Use `/visual-edit` when a UI needs to be reviewed or changed
  169. in context, including responsive states and multi-screen flows.
  170. Install the skill and Design connector from your app repository:
  171. ```sh
  172. npx @agent-native/core@latest skills add visual-edit
  173. ```
  174. Try it on one screen, several routes, a complete flow, or responsive sizes:
  175. - `/visual-edit the dashboard at desktop size`
  176. - `/visual-edit compare the dashboard, settings, and reports pages side by side`
  177. - `/visual-edit the onboarding flow from welcome through account created`
  178. - `/visual-edit show the dashboard at desktop, tablet, and mobile widths`
  179. Read the [public Visual Edit README](./skills/visual-edit/README.md) for a
  180. hands-on example, and the [Agent-Native guide](https://github.com/BuilderIO/agent-native/blob/main/skills/visual-edit/README.md)
  181. for the full workflow.
  182. ### [`/remix`](skills/remix/README.md)
  183. Take a section, effect, or interaction from one website and rebuild it in your
  184. own site. Design puts the source and your site side by side; select what to take
  185. and where it goes (both at once is fine), and the agent reads the source's
  186. shipped code, rebuilds it with your colors, fonts, and components, and keeps
  187. images, video, and 3D models swappable.
  188. Every shader, WebGL, 3D, or animation effect gets knobs in the Design
  189. inspector's Tweaks section, shown whenever the effect or anything around it is
  190. selected. Your agent can read and set the same values over MCP.
  191. `/remix explain` lays out a small explainer beside the source instead: one live,
  192. tweakable demo per idea.
  193. Install the skill and Design connector from your app repository:
  194. ```sh
  195. npx @agent-native/core@latest skills add remix
  196. ```
  197. Try:
  198. - `/remix the hero background from stripe.com into my homepage`
  199. - `/remix the pricing table from linear.app into /pricing`
  200. - `/remix explain the scroll animation on this page`
  201. Read the [public Remix README](./skills/remix/README.md) for a worked example.
  202. ### [`/rewind`](skills/rewind/README.md)
  203. Use local Clips Rewind screen memory to recover a recent moment: what was said,
  204. what appeared on screen, or what happened just before the current task. It
  205. searches bounded local chapters, transcripts, OCR, and frames first; it is not a
  206. hosted recording archive and does not upload raw local media by default.
  207. Rewind requires macOS, the signed
  208. [Clips Desktop app](https://clips.agent-native.com/download), and a compatible
  209. agent with the local `clips-screen-memory` MCP connection. If you invoke Rewind
  210. before installing Clips, the skill asks permission to open the official install
  211. flow instead of silently installing capture software. Follow the [Rewind setup
  212. and first-test guide](skills/rewind/README.md).
  213. ### [`/visual-plan`](skills/visual-plan/README.md)
  214. Turn ordinary text plans into rich interactive visual plans with diagrams, file
  215. maps, annotated code, open questions, and UI/prototype review when useful.
  216. Solves for plans that are too important to bury in chat. The output is
  217. scannable, commentable, and intuitive enough for a human to approve before code
  218. changes start.
  219. <picture>
  220. <img alt="Visual plan review surface" src="media/visual-plan.png">
  221. </picture>
  222. Visual plans are MDX, customizable with your own components, and are viewed with the [Agent-Native plans app](https://www.agent-native.com/docs/template-plan). In local-files mode, `/visual-plan` writes and serves MDX locally through a localhost bridge instead of uploading plan content to the hosted database. [Source here](https://github.com/BuilderIO/agent-native/)
  223. ### [`/visual-recap`](skills/visual-recap/README.md)
  224. Turn a branch, commit, or PR diff into an interactive visual recap with
  225. annotated diffs, diagrams, API/schema summaries, file maps, UI state summaries,
  226. and focused review notes.
  227. Solves for diffs that hide the shape of the change. Reviewers can understand
  228. contracts, architecture moves, schema changes, and UI impact before diving into
  229. raw line-by-line review.
  230. <picture>
  231. <img alt="Visual recap review surface animation" src="media/visual-recap.gif">
  232. </picture>
  233. Visual recaps are MDX, customizable with your own components, and are viewed with the [Agent-Native plans app](https://www.agent-native.com/docs/template-plan). [Source here](https://github.com/BuilderIO/agent-native/)
  234. You can also install a GitHub action for these to be automatically generated for every PR with
  235. ```sh
  236. npx @agent-native/skills@latest add
  237. ```
  238. ![Example of a visual plan posted to a PR](https://cdn.builder.io/api/v1/image/assets%2FYJIGb4i01jvw0SRdL5Bt%2Fcf9bac396cf24a4ba976fc331af6fc5d)
  239. ### [`/agent-watchdog`](skills/agent-watchdog/README.md)
  240. Audit another agent's work from a Codex session, Claude Code transcript, PR,
  241. branch, or run summary.
  242. Solves for cross-agent handoffs: watch until done, reconstruct what was asked,
  243. check what actually changed and verified, report gaps, and optionally make
  244. narrow fixes.
  245. ### [`/plan-arbiter`](skills/plan-arbiter/README.md)
  246. Compare competing agent plans and choose one executable direction.
  247. Solves for multi-agent planning loops where Codex, Claude Code, or other agents
  248. produce separate strategies. The output is a decision memo with the winning or
  249. hybrid plan, rejected alternatives, verification gates, and executor
  250. recommendation.
  251. ### [`/plow-ahead`](skills/plow-ahead/README.md)
  252. Keep working through ordinary ambiguity and finish with a clear decision recap.
  253. Solves for explicit autonomy requests: the agent converts routine questions into
  254. assumptions, proceeds with conservative choices, validates the work, and recaps
  255. the decisions it made without stopping.
  256. ### [`/efficient-orchestrator`](skills/efficient-orchestrator/README.md)
  257. Run an expensive model (Claude Fable or Opus, OpenAI Astra or Sol) as the
  258. orchestrator, architect and final judge while cheaper agents (Haiku or Sonnet,
  259. Terra or Luna) handle token-heavy research, coding, QA and log reduction. It runs
  260. only when invoked by name; `/efficient-frontier` and `/efficient-fable` are
  261. aliases.
  262. Solves for expensive-model waste without unrequested fan-out: the premium model
  263. spends tokens on judgment, and helpers stay capped and short-lived.
  264. <picture>
  265. <source media="(prefers-color-scheme: dark)" srcset="skills/efficient-orchestrator/assets/fable-orchestrator-dark.png">
  266. <source media="(prefers-color-scheme: light)" srcset="skills/efficient-orchestrator/assets/fable-orchestrator.png">
  267. <img alt="Orchestrator diagram" src="skills/efficient-orchestrator/assets/fable-orchestrator.png">
  268. </picture>
  269. ### [`/stay-within-limits`](skills/stay-within-limits/README.md)
  270. Check current 5-hour and weekly usage before substantial work and between
  271. parallel waves, then pause new execution at 95% until the active window is clear
  272. enough to continue.
  273. Solves for long-running agent sessions that accidentally exhaust the current
  274. budget window mid-task instead of pausing cleanly and resuming with a
  275. self-contained plan.
  276. ### [`/quick-recap`](skills/quick-recap/README.md)
  277. Add a concise final status block convention so every completed response ends
  278. with a clear green, yellow, or red work-state signal.
  279. Solves for ambiguity at the end of agent work: done, pending a specific
  280. non-routine step, or blocked on the user.
  281. Example green status:
  282. ```md
  283. 🟢 Updated quick recap docs with output examples
  284. ```
  285. Example yellow status:
  286. ```md
  287. 🟡 Code updated, set PROVIDER_WEBHOOK_SECRET before testing webhooks
  288. ```
  289. ### [`/read-the-damn-docs`](skills/read-the-damn-docs/README.md)
  290. Make agents web-search for authoritative docs before they guess from stale model
  291. memory.
  292. Solves for version drift and API folklore: package installs, framework config,
  293. SDK imports, provider limits, auth, security, billing, data, migrations, deploys,
  294. and repo-specific contracts all require a docs pass before implementation. For
  295. external APIs and current product behavior, web search for official docs is
  296. usually the first move.
  297. ### [`/turn-into-app`](skills/turn-into-app/README.md)
  298. Give a workflow you have already proven the face of an app and the brain of an
  299. agent. Turn a thread, a skill, a spreadsheet, or a Claude or ChatGPT project
  300. into a visual Agent-Native app that opens on your world, populated: a week grid
  301. for scheduling work, a board for stages, a live workbench for a spreadsheet
  302. model, a triage queue with a drafted reply. Buttons sit on the objects they act
  303. on, so you click, watch the agent work, and see the result land back in the app.
  304. A form under a step bar is the failure it is built to avoid. The skill picks the
  305. app's shape from the source, commits to a visual direction with real light and
  306. dark tokens, seeds realistic sample data, builds the agent moments into the
  307. objects, and reviews its own screenshots at desktop and phone sizes before it
  308. hands the app back.
  309. <picture>
  310. <img alt="Before and after: a form and a plain table become a findings workbench, a spreadsheet workbench with sliders and a chart, and a call-prep queue" src="media/turn-into-app-before-after.png">
  311. </picture>
  312. Install it, then try one of these:
  313. ```sh
  314. npx @agent-native/skills@latest add --skill turn-into-app
  315. ```
  316. - `/turn-into-app` at the end of a thread
  317. - `/turn-into-app /some-skill` to package a skill
  318. - `/turn-into-app ./forecast.xlsx` to turn a spreadsheet into a workbench
  319. - `/turn-into-app ./project-export/` to turn a Claude or ChatGPT project into an app
  320. It builds and verifies locally in Claude Code, Codex, Cursor, and other coding
  321. agents; Claude and ChatGPT on the web hand a bounded brief to Builder through
  322. Dispatch. Read the [full guide](skills/turn-into-app/README.md).
  323. ## Install
  324. Run the installer:
  325. ```sh
  326. npx @agent-native/skills@latest add
  327. ```
  328. The picker shows the full catalog, with the recommended skills preselected.
  329. Toggle any additional skills you want.
  330. The installer walks you through the choices:
  331. - Which skills to install.
  332. - Where visual plans and recaps should live: hosted shareable links
  333. (recommended), local files only, or a self-hosted/custom Plan app.
  334. - Agent Skills through the shared `.agents` path for Codex, Pi, Cursor,
  335. OpenCode, GitHub Copilot / VS Code, and similar agents, plus Claude Code's
  336. native skills path when selected.
  337. - User-level or project-level install.
  338. - Whether to add managed `AGENTS.md` / `CLAUDE.md` instruction blocks when the
  339. selected skills have always-on guidance.
  340. - Whether to add the PR Visual Recap GitHub Action when `/visual-recap` is
  341. selected.
  342. - Whether to configure a compatible local agent connection when `/rewind` is
  343. selected. The installer does not install Clips Desktop. If it is missing,
  344. invoking Rewind asks permission to open the official download flow; capture
  345. remains disabled until you turn it on in the Clips tray.
  346. Skip the picker with `--skill`:
  347. ```sh
  348. npx @agent-native/skills@latest add --skill quick-recap
  349. npx @agent-native/skills@latest add --skill visual-recap --with-github-action
  350. npx @agent-native/skills@latest add --skill rewind
  351. npx @agent-native/skills@latest add --skill webmcp
  352. ```
  353. You can also use Vercel's `skills` CLI for a plain skill-folder copy:
  354. ```sh
  355. npx skills@latest add BuilderIO/skills --skill quick-recap
  356. ```
  357. Do not use the plain-copy installer for Rewind: it cannot configure the local
  358. `clips-screen-memory` connection. Use `@agent-native/skills` or
  359. `@agent-native/core` for a complete Rewind setup.
  360. That installer is useful for quick copying, but it does not add the managed
  361. `AGENTS.md` / `CLAUDE.md` instruction blocks or the PR Visual Recap GitHub
  362. Action that pair well with these skills.
  363. ### Install in Claude Cowork
  364. This repo is also a Claude plugin marketplace for Cowork. To install the
  365. shareable, updatable plugin in Cowork:
  366. 1. Open `Customize` > `Plugins` > `Add marketplace`.
  367. 2. Enter `BuilderIO/skills` (or `https://github.com/BuilderIO/skills`).
  368. 3. Install `Builder Skills` and enable the `webmcp` skill.
  369. Invoke the installed skill with its plugin namespace:
  370. ```text
  371. /builder-skills:webmcp slides inspect the current screen without editing
  372. ```
  373. For Team and Enterprise workspaces, an organization admin can provision the
  374. marketplace or plugin for everyone. Cowork does not currently expose a
  375. supported CLI command for installing an account-level plugin; the marketplace
  376. is shared through Git, while each Cowork user installs it from `Customize`.
  377. ### Install in Claude Code
  378. This repo is also a [Claude Code plugin
  379. marketplace](https://code.claude.com/docs/en/plugin-marketplaces). To install
  380. all of the skills as a managed, updatable plugin, run these inside Claude Code:
  381. ```sh
  382. /plugin marketplace add BuilderIO/skills
  383. /plugin install builder-skills@builder-skills
  384. ```
  385. The skills are then namespaced under the plugin (for example,
  386. `/builder-skills:quick-recap`). Pull future updates with:
  387. ```sh
  388. /plugin marketplace update builder-skills
  389. ```
  390. The plugin installs Rewind's instructions only; it cannot configure the local
  391. Clips Screen Memory MCP connection. When invoked, Rewind can ask permission to
  392. open the official Clips Desktop installer, but it must not install or enable
  393. capture silently. After Clips is installed and Rewind is enabled, also run:
  394. ```sh
  395. npx @agent-native/core@latest skills add rewind --client claude-code --scope user --yes
  396. ```
  397. Treat Rewind as unavailable until `screen_memory_status` succeeds.
  398. This path does not add the managed `AGENTS.md` / `CLAUDE.md` instruction blocks
  399. or the PR Visual Recap GitHub Action; use the `npx @agent-native/skills`
  400. installer above if you want those.

Ratings and comments

Sign in to rate this skill.

Sign in

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