![\"Memi](\"https://raw.githubusercontent.com/memi-design/memi/main/assets/memi-brand-banner.png\")
\n\n \n
\n ![\"npm](\"https://img.shields.io/npm/v/@memi-design/cli?color=bd3f63&label=npm\")
![\"weekly](\"https://img.shields.io/npm/dw/@memi-design/cli?color=171718&label=weekly%20downloads\")
![\"CI\"](\"https://github.com/memi-design/memi/actions/workflows/ci.yml/badge.svg?branch=main\"){kind=icon}
\n ![\"GitHub](\"https://img.shields.io/github/stars/memi-design/memi?style=social\")
![\"MIT](\"https://img.shields.io/badge/license-MIT-171718.svg\"){kind=icon}
\n
\n Start with your next interface ·\n Get Memi Studio ·\n Read the research\n
\n\n## Quickstart: find your first interface issue\n\nRun one non-destructive audit in any frontend repository. It needs no account, API key, Figma file, global install, or daemon.\n\n```bash\nnpx -y @memi-design/cli@latest diagnose . --json --no-write --fail-on none\n```\n\nThe result carries normalized finding IDs, confidence, provenance, and `file:line` evidence so an agent can act on a specific finding instead of guessing.\n\nGive the same context to your coding agent:\n\n```bash\nnpx skills add memi-design/memi --skill audit-frontend-design\n```\n\nThen ask:\n\n> Audit this frontend before editing it. Prioritize the five changes that will matter most to users, reuse the existing system, and verify the result after the patch.\n\nIf Memi catches a real interface issue in your project, [share the finding](https://github.com/memi-design/memi/discussions/categories/show-and-tell). Real reports are the most useful signal for what to improve next.\n\n## Put the check on every pull request\n\nCopy [`examples/github-actions/memi-design.yml`](examples/github-actions/memi-design.yml) into your repository as `.github/workflows/memi-design.yml`. The starter is pinned to the reviewed public Action commit and gives reviewers:\n\n- a PR check that fails only on newly introduced interface debt;\n- a `memi-design-health` artifact with the human-readable report; and\n- SARIF annotations when the repository grants `security-events: write`.\n\nThe workflow does not need an API key or a Memi secret. Fork pull requests still receive the check and report; SARIF upload is skipped automatically when GitHub does not grant that permission.\n\nIf you prefer to configure it by hand, the complete [GitHub Action guide](https://github.com/memi-design/memi/blob/main/docs/GITHUB_ACTION_MARKETPLACE.md) documents every input, output, permission, and evidence file.\n\n## One product layer, three surfaces\n\n| Surface | What it is | Status |\n| --- | --- | --- |\n| **Memi CLI** | Interface intelligence and deterministic checks for local repositories, agents, and CI. | Available today |\n| **Memi Studio** | A macOS workbench for bringing project context, agent workflows, and verification together. | [Available today](https://memoire.cv/download) |\n| **Memi Canvas** | A visual workspace for design-system context and controlled agent proposals. | currently in development |\n\n## See the product\n\n| Memi Studio | Memi Canvas — in development |\n| --- | --- |\n|![\"Memi](\"https://raw.githubusercontent.com/memi-design/memi/main/product-hunt-assets/source-captures-v2/studio-real-01-workbench.png\")
| ![\"Memi](\"https://raw.githubusercontent.com/memi-design/memi/main/assets/product/memi-canvas-workspace.png\")
|\n| Bring an agent prompt, project memory, and a verification surface into one workbench. | Preview design-system context, inspect a proposal, and keep a human in the loop. This preview shows an active development build, not a released product guarantee. |\n\n## What Memi adds to an agent workflow\n\n| Before the edit | During the edit | Before merge |\n| --- | --- | --- |\n| Discover components, tokens, routes, states, and accessibility gaps. | Give the agent a scoped brief that names the system it must preserve. | Rerun deterministic checks and surface new interface debt in CI. |\n\n| Need | Start with |\n| --- | --- |\n| Find UI risks and product-system context | `audit-frontend-design` |\n| Plan a change around existing components and tokens | `remember-design-system` |\n| Keep new interface debt out of pull requests | `enforce-design-ci` |\n| Build and verify native Apple interfaces | `build-swiftui-interface` |\n\nCompatible with the [shadcn registry](https://ui.shadcn.com/docs/registry/getting-started) and [v0 design systems](https://v0.app/docs/design-systems).\n\n## Evidence at a glance\n\nThe [V15 confirmatory audit](https://github.com/memi-design/memi/tree/main/docs/research/memi-2.7-prospective-study/v15-2.7.3-confirmatory) is a public technical disclosure, not a leaderboard. It separates receipt admission, rendered design quality, functional acceptance, and resource observations.\n\n| Measured record | Exact reading |\n| --- | --- |\n| 36 / 36 frozen receipts admitted | Every preregistered agent cell had an auditable receipt. This is receipt admission, not universal performance. |\n| **10 complete model-graded matched pairs** | Rendered design-quality comparisons that survived the prespecified screen. This is model-graded evidence, not independent practitioner review. |\n| Buzzr / Expo: mean **+1.4**; Paraform / web: mean **−0.4** | The scoped non-inferiority gate passed on both graded task families. It does not establish general superiority. |\n| **0 / 21** corrected task-by-resource tests rejected | The study did not establish a speed, cost, or token-use advantage. |\n\n**Separate historical release record:** the 2.7 candidate record reported **2,187 / 2,187** tests passed. It is release evidence, not part of V15 and not proof that every project benefits.\n\n## Benchmarks and paper\n\n\n \n ![\"V15](\"https://raw.githubusercontent.com/memi-design/memi/main/assets/readme-benchmark.svg\")
\n \n