[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6R6AJZ4MMXR6K89WZXAAK4Q":3,"skill-same:01M6R6AJZ4MMXR6K89WZXAAK4Q":71},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":25,"publishedAt":46,"linkOnly":47,"markdownContent":48,"markdownBody":49,"frontMatter":50,"refName":51,"commitSha":52,"skillDirectory":53,"forksCount":13,"repositoryHtmlUrl":54,"repositoryHomepageUrl":6,"ownerHtmlUrl":55,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":56,"install":57,"resources":70},"01M6R6AJZ4MMXR6K89WZXAAK4Q","video-test",null,"Record a video test of any feature or user flow via Playwright. Use when the user says \"record video test\", \"video test\", \"record a test\", \"show me the feature\", \"record the UI\", \"capture video\", or wants a visual demo of a feature. Accepts a description of what to test as an argument (e.g., `\u002Fvideo-test the brainstorm creation flow`).","diegoquinteiro\u002Fharnessfile","diegoquinteiro","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F1878108?v=4","01M6R66DV4959S3MJ0BSC20JMG","01M6R66DYYMACQCANZQ4DFGEPR",0,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fdiegoquinteiro\u002Fharnessfile\u002Fblob\u002Fmain\u002Fui\u002Fsrc\u002Fdemo\u002Faltavox\u002Fskills\u002Fvideo-test\u002FSKILL.md",[17,22],{"categoryNo":18,"categoryName":19,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000008","测试与质量","SCENARIO","使用场景",{"categoryNo":23,"categoryName":24,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000016","图像与多媒体",[26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45],"test","video test","Video","me","playwright","when","argument","via","as","feature","creation","visual","demo","record","show","use","brainstorm","show me","record video","flow","2026-11-10 10:12:30",false,"---\nname: video-test\ndescription: Record a video test of any feature or user flow via Playwright. Use when the user says \"record video test\", \"video test\", \"record a test\", \"show me the feature\", \"record the UI\", \"capture video\", or wants a visual demo of a feature. Accepts a description of what to test as an argument (e.g., `\u002Fvideo-test the brainstorm creation flow`).\n---\n\n# Video Test\n\nWrite and run a Playwright browser test with video recording for any feature or user flow, then attach the resulting video to the PR for the current branch.\n\n## Step 1: Understand what to test\n\n- The user provides a description of the feature\u002Fflow to test via the skill argument or conversation context.\n- If no description is provided, ask the user what they'd like to test.\n- Explore the relevant frontend routes in `web\u002Fapp\u002F` and backend GraphQL schemas to understand the flow and the pages involved.\n- This can be any feature or flow — not just what's being developed in the current branch.\n\n## Step 2: Start the app stack\n\nStart the full stack (backend + frontend + database) in E2E mode with the repo's run-app\nscript (a checked-in script — no skill dependency):\n\n```bash\nscripts\u002Frun-app.sh --skip-tunnels\n```\n\nThe script outputs the ports and RUN_ID. Parse them from the output:\n- `RUN_ID` — needed for cleanup\n- `FRONTEND_PORT` — the port the frontend is running on\n- `BACKEND_PORT` — the port the backend is running on\n\nEnsure Playwright and Chromium are installed in `e2e\u002F`:\n```bash\ncd e2e && npm install && npx playwright install chromium\n```\n\n### How E2E auth works\n\nThe `\u002Frun-app` script runs in E2E mode by default:\n- **Frontend**: Clerk auth is bypassed, Apollo sends `X-E2E-User-Id`\u002F`X-E2E-Org-Id` headers\n- **Backend**: E2E auth middleware recognizes headers and sets user\u002Faccount context\n\nNo Clerk login flow is needed. All authenticated pages are accessible.\n\n## Step 3: Write the Playwright test\n\nCreate `e2e\u002Ftests\u002F\u003Cfeature>.video.spec.ts`. Key requirements:\n\n- Import `test` and `expect` from `@playwright\u002Ftest`\n- Use the `page` fixture (browser context), NOT the `request` fixture\n- **Name the file `\u003Cfeature>.video.spec.ts`** (the `.video.spec.ts` suffix matters — see below).\n- **Do NOT hard-code `video.size` via `test.use`.** Video size and viewport come from the Playwright `projects` (`desktop` 1280×720 + `mobile` iPhone 13) in `e2e\u002Fplaywright.config.ts`. The `mobile` project is scoped (via `testMatch`) to `*.video.spec.ts` only, so a correctly-named video spec runs once per project — producing a desktop and a mobile video automatically — while the rest of the suite stays desktop-only. If you need a per-run `baseURL`, set only that:\n  ```typescript\n  test.use({ baseURL: \"http:\u002F\u002Flocalhost:\u003CFRONTEND_PORT>\" });\n  ```\n- **Pre-warm the route before the watchable part** to avoid the long white intro: in dev (`yarn dev`) Next compiles the route on first hit, which is the main cause of the blank video start. Begin the test by navigating and waiting for the UI to actually mount, *before* any deliberate pause:\n  ```typescript\n  await page.goto(\"\u002Fpath\");\n  await page.waitForLoadState(\"networkidle\");\n  await expect(page.getByRole(\"main\")).toBeVisible(); \u002F\u002F an anchor that proves the UI mounted\n  ```\n- **Use `{ force: true }` on click actions** — the app has overlay elements (sidebar, page containers) that can intercept pointer events\n- Add deliberate pauses (`await page.waitForTimeout(2000)`) at key moments so the video is watchable\n- **Capture anchor screenshots (3–4 per viewport)** at the key moments of the flow, so reviewers get quick visual evidence without watching the video. Prefix the filename with the project name so desktop and mobile don't collide, and save under a predictable per-project folder:\n  ```typescript\n  const shotDir = `test-results\u002F${test.info().project.name}\u002Fscreenshots`;\n  \u002F\u002F (1) first useful screen, (2) main action, (3) result\u002Ffinal state, (4) optional confirmation\n  await page.screenshot({ path: `${shotDir}\u002F01-first-screen.png` });\n  \u002F\u002F ...later, after the main interaction:\n  await page.screenshot({ path: `${shotDir}\u002F02-main-action.png` });\n  \u002F\u002F ...at the result\u002Ffinal state:\n  await page.screenshot({ path: `${shotDir}\u002F03-result.png` });\n  ```\n- Use try\u002Fcatch around actions that might fail (e.g., dropdown selections, navigation waits) to keep the video recording going\n- Use descriptive test names that explain what is being demonstrated\n\nThe test file should be a **real integration E2E test** that will be committed to the project under `e2e\u002Ftests\u002F`. Name it descriptively based on the feature **with the `.video.spec.ts` suffix** (e.g., `brainstorm-creation.video.spec.ts`, `search-flow.video.spec.ts`). The suffix is what opts the spec into the dual desktop+mobile recording (the `mobile` project's `testMatch`); a plain `*.spec.ts` name would only record desktop. Do NOT use generic names like `video-recording.video.spec.ts`. The test should have proper assertions that verify the feature actually works — do NOT use try\u002Fcatch to swallow failures.\n\n## Step 4: Run the test (desktop + mobile)\n\nRun the spec across **both** viewports (the `desktop` and `mobile` projects defined in `playwright.config.ts`):\n\n```bash\ncd e2e && npx playwright test tests\u002F\u003Cfeature>.video.spec.ts\n```\n\nThis produces one video per project. Playwright saves them under\n`e2e\u002Ftest-results\u002F\u003C...>-\u003Cproject>\u002Fvideo.webm` and the screenshots under\n`e2e\u002Ftest-results\u002F\u003Cproject>\u002Fscreenshots\u002F`. To record a single viewport, add\n`--project=desktop` or `--project=mobile`.\n\nThe video is saved even if the test fails. If the test fails, review the error\nand the error-context.md file (contains a page snapshot). Fix the test and\nre-run. Iterate until the test passes and produces a good video on both\nviewports.\n\n## Step 5: Trim the white intro from each video (ffmpeg)\n\nEven with the route pre-warmed, Playwright starts recording when the browser\ncontext is created — a fraction of a second before the first useful frame. Trim\nthat residual head off each `.webm` so the video opens on real UI:\n\n```bash\nfor VIDEO in $(find e2e\u002Ftest-results -name \"*.webm\" -type f); do\n  TRIMMED=\"${VIDEO%.webm}.trimmed.webm\"\n  # Simple, robust: drop a conservative fixed head (~1.5s).\n  ffmpeg -y -ss 1.5 -i \"$VIDEO\" -c copy \"$TRIMMED\" && mv \"$TRIMMED\" \"$VIDEO\"\ndone\n```\n\nIf a fixed offset still leaves white (e.g. a slow first compile), detect where\nthe white intro ends and cut from there. The intro is **white** (luma ~235), so\n`blackdetect` alone won't catch it — `negate` first so white becomes black, then\n`blackdetect`'s `black_end` marks the moment the UI appears:\n\n```bash\nSTART=$(ffmpeg -i \"$VIDEO\" -vf \"negate,blackdetect=d=0.1:pic_th=0.98\" -an -f null - 2>&1 \\\n  | grep -oP 'black_end:\\K[0-9.]+' | head -1)\nffmpeg -y -ss \"${START:-1.5}\" -i \"$VIDEO\" -c copy \"$VIDEO.trimmed.webm\"\n```\n\n## Step 6: Attach key screenshots to the Multica issue\n\nThe anchor screenshots go on the **issue** (where humans review), while the\nvideo goes on the PR (Step 7). Collect the PNGs and attach them in a single\ncomment. The issue ID comes from the task context (e.g. `ALT-147`'s UUID); if it\nis not available, skip this step and just report the local screenshot paths.\n\n```bash\nSHOTS=$(find e2e\u002Ftest-results -path '*\u002Fscreenshots\u002F*.png' -type f | sort)\nATTACH=$(printf -- '--attachment %q ' $SHOTS)\nmultica issue comment add \u003CISSUE_ID> \\\n  --content \"Visual E2E evidence (desktop + mobile) — key moments of the flow.\" \\\n  $ATTACH\n```\n\n## Step 7: Attach video to PR\n\n1. Find the video files (one per viewport):\n   ```bash\n   VIDEO_PATHS=$(find e2e\u002Ftest-results -name \"*.webm\" -type f | sort)\n   ```\n\n2. Upload each .webm directly to catbox.moe (no GIF conversion needed — GitHub renders video links):\n   ```bash\n   VIDEO_URL=$(curl -s -F \"reqtype=fileupload\" -F \"fileToUpload=@$VIDEO_PATH\" https:\u002F\u002Fcatbox.moe\u002Fuser\u002Fapi.php)\n   ```\n   Do this for both the desktop and mobile videos and keep both URLs.\n\n3. Get the current branch and PR:\n   ```bash\n   PR_JSON=$(gh pr view --json number,url 2>\u002Fdev\u002Fnull)\n   ```\n\n   If no PR exists, create one:\n   ```bash\n   BRANCH=$(git branch --show-current)\n   gh pr create --title \"$BRANCH\" --body \"## Summary\n\n   (auto-created by video-test skill)\n\n   ? Generated with [Claude Code](https:\u002F\u002Fclaude.com\u002Fclaude-code)\"\n   PR_JSON=$(gh pr view --json number,url)\n   ```\n\n4. Post a PR comment with **both** videos linked:\n   ```bash\n   PR_NUMBER=$(echo \"$PR_JSON\" | jq -r '.number')\n   gh pr comment \"$PR_NUMBER\" --body \"## Video Test\n\n   ▶️ [Desktop recording]($DESKTOP_VIDEO_URL)\n   ▶️ [Mobile recording]($MOBILE_VIDEO_URL)\n\n   Recorded with Playwright on $(date -u +%Y-%m-%dT%H:%M:%SZ)\"\n   ```\n\n5. **Fallback** if catbox upload fails: Tell the user the local path to the .webm file and suggest they drag-and-drop it into the PR on GitHub's web UI. On macOS: `open \u003Cpath>` to preview.\n\n## Step 8: Cleanup\n\n```bash\nscripts\u002Fstop-app.sh $RUN_ID\n```\n\n## Step 9: Report results\n\n- Confirm whether the videos (desktop + mobile) were attached to the PR and the screenshots to the issue\n- Include the PR URL\n- Report the local video\u002Fscreenshot paths as backup\n- **Return the VIDEO_URLs** so callers can attach them to the Multica issue or the PR\n","# Video Test\n\nWrite and run a Playwright browser test with video recording for any feature or user flow, then attach the resulting video to the PR for the current branch.\n\n## Step 1: Understand what to test\n\n- The user provides a description of the feature\u002Fflow to test via the skill argument or conversation context.\n- If no description is provided, ask the user what they'd like to test.\n- Explore the relevant frontend routes in `web\u002Fapp\u002F` and backend GraphQL schemas to understand the flow and the pages involved.\n- This can be any feature or flow — not just what's being developed in the current branch.\n\n## Step 2: Start the app stack\n\nStart the full stack (backend + frontend + database) in E2E mode with the repo's run-app\nscript (a checked-in script — no skill dependency):\n\n```bash\nscripts\u002Frun-app.sh --skip-tunnels\n```\n\nThe script outputs the ports and RUN_ID. Parse them from the output:\n- `RUN_ID` — needed for cleanup\n- `FRONTEND_PORT` — the port the frontend is running on\n- `BACKEND_PORT` — the port the backend is running on\n\nEnsure Playwright and Chromium are installed in `e2e\u002F`:\n```bash\ncd e2e && npm install && npx playwright install chromium\n```\n\n### How E2E auth works\n\nThe `\u002Frun-app` script runs in E2E mode by default:\n- **Frontend**: Clerk auth is bypassed, Apollo sends `X-E2E-User-Id`\u002F`X-E2E-Org-Id` headers\n- **Backend**: E2E auth middleware recognizes headers and sets user\u002Faccount context\n\nNo Clerk login flow is needed. All authenticated pages are accessible.\n\n## Step 3: Write the Playwright test\n\nCreate `e2e\u002Ftests\u002F\u003Cfeature>.video.spec.ts`. Key requirements:\n\n- Import `test` and `expect` from `@playwright\u002Ftest`\n- Use the `page` fixture (browser context), NOT the `request` fixture\n- **Name the file `\u003Cfeature>.video.spec.ts`** (the `.video.spec.ts` suffix matters — see below).\n- **Do NOT hard-code `video.size` via `test.use`.** Video size and viewport come from the Playwright `projects` (`desktop` 1280×720 + `mobile` iPhone 13) in `e2e\u002Fplaywright.config.ts`. The `mobile` project is scoped (via `testMatch`) to `*.video.spec.ts` only, so a correctly-named video spec runs once per project — producing a desktop and a mobile video automatically — while the rest of the suite stays desktop-only. If you need a per-run `baseURL`, set only that:\n  ```typescript\n  test.use({ baseURL: \"http:\u002F\u002Flocalhost:\u003CFRONTEND_PORT>\" });\n  ```\n- **Pre-warm the route before the watchable part** to avoid the long white intro: in dev (`yarn dev`) Next compiles the route on first hit, which is the main cause of the blank video start. Begin the test by navigating and waiting for the UI to actually mount, *before* any deliberate pause:\n  ```typescript\n  await page.goto(\"\u002Fpath\");\n  await page.waitForLoadState(\"networkidle\");\n  await expect(page.getByRole(\"main\")).toBeVisible(); \u002F\u002F an anchor that proves the UI mounted\n  ```\n- **Use `{ force: true }` on click actions** — the app has overlay elements (sidebar, page containers) that can intercept pointer events\n- Add deliberate pauses (`await page.waitForTimeout(2000)`) at key moments so the video is watchable\n- **Capture anchor screenshots (3–4 per viewport)** at the key moments of the flow, so reviewers get quick visual evidence without watching the video. Prefix the filename with the project name so desktop and mobile don't collide, and save under a predictable per-project folder:\n  ```typescript\n  const shotDir = `test-results\u002F${test.info().project.name}\u002Fscreenshots`;\n  \u002F\u002F (1) first useful screen, (2) main action, (3) result\u002Ffinal state, (4) optional confirmation\n  await page.screenshot({ path: `${shotDir}\u002F01-first-screen.png` });\n  \u002F\u002F ...later, after the main interaction:\n  await page.screenshot({ path: `${shotDir}\u002F02-main-action.png` });\n  \u002F\u002F ...at the result\u002Ffinal state:\n  await page.screenshot({ path: `${shotDir}\u002F03-result.png` });\n  ```\n- Use try\u002Fcatch around actions that might fail (e.g., dropdown selections, navigation waits) to keep the video recording going\n- Use descriptive test names that explain what is being demonstrated\n\nThe test file should be a **real integration E2E test** that will be committed to the project under `e2e\u002Ftests\u002F`. Name it descriptively based on the feature **with the `.video.spec.ts` suffix** (e.g., `brainstorm-creation.video.spec.ts`, `search-flow.video.spec.ts`). The suffix is what opts the spec into the dual desktop+mobile recording (the `mobile` project's `testMatch`); a plain `*.spec.ts` name would only record desktop. Do NOT use generic names like `video-recording.video.spec.ts`. The test should have proper assertions that verify the feature actually works — do NOT use try\u002Fcatch to swallow failures.\n\n## Step 4: Run the test (desktop + mobile)\n\nRun the spec across **both** viewports (the `desktop` and `mobile` projects defined in `playwright.config.ts`):\n\n```bash\ncd e2e && npx playwright test tests\u002F\u003Cfeature>.video.spec.ts\n```\n\nThis produces one video per project. Playwright saves them under\n`e2e\u002Ftest-results\u002F\u003C...>-\u003Cproject>\u002Fvideo.webm` and the screenshots under\n`e2e\u002Ftest-results\u002F\u003Cproject>\u002Fscreenshots\u002F`. To record a single viewport, add\n`--project=desktop` or `--project=mobile`.\n\nThe video is saved even if the test fails. If the test fails, review the error\nand the error-context.md file (contains a page snapshot). Fix the test and\nre-run. Iterate until the test passes and produces a good video on both\nviewports.\n\n## Step 5: Trim the white intro from each video (ffmpeg)\n\nEven with the route pre-warmed, Playwright starts recording when the browser\ncontext is created — a fraction of a second before the first useful frame. Trim\nthat residual head off each `.webm` so the video opens on real UI:\n\n```bash\nfor VIDEO in $(find e2e\u002Ftest-results -name \"*.webm\" -type f); do\n  TRIMMED=\"${VIDEO%.webm}.trimmed.webm\"\n  # Simple, robust: drop a conservative fixed head (~1.5s).\n  ffmpeg -y -ss 1.5 -i \"$VIDEO\" -c copy \"$TRIMMED\" && mv \"$TRIMMED\" \"$VIDEO\"\ndone\n```\n\nIf a fixed offset still leaves white (e.g. a slow first compile), detect where\nthe white intro ends and cut from there. The intro is **white** (luma ~235), so\n`blackdetect` alone won't catch it — `negate` first so white becomes black, then\n`blackdetect`'s `black_end` marks the moment the UI appears:\n\n```bash\nSTART=$(ffmpeg -i \"$VIDEO\" -vf \"negate,blackdetect=d=0.1:pic_th=0.98\" -an -f null - 2>&1 \\\n  | grep -oP 'black_end:\\K[0-9.]+' | head -1)\nffmpeg -y -ss \"${START:-1.5}\" -i \"$VIDEO\" -c copy \"$VIDEO.trimmed.webm\"\n```\n\n## Step 6: Attach key screenshots to the Multica issue\n\nThe anchor screenshots go on the **issue** (where humans review), while the\nvideo goes on the PR (Step 7). Collect the PNGs and attach them in a single\ncomment. The issue ID comes from the task context (e.g. `ALT-147`'s UUID); if it\nis not available, skip this step and just report the local screenshot paths.\n\n```bash\nSHOTS=$(find e2e\u002Ftest-results -path '*\u002Fscreenshots\u002F*.png' -type f | sort)\nATTACH=$(printf -- '--attachment %q ' $SHOTS)\nmultica issue comment add \u003CISSUE_ID> \\\n  --content \"Visual E2E evidence (desktop + mobile) — key moments of the flow.\" \\\n  $ATTACH\n```\n\n## Step 7: Attach video to PR\n\n1. Find the video files (one per viewport):\n   ```bash\n   VIDEO_PATHS=$(find e2e\u002Ftest-results -name \"*.webm\" -type f | sort)\n   ```\n\n2. Upload each .webm directly to catbox.moe (no GIF conversion needed — GitHub renders video links):\n   ```bash\n   VIDEO_URL=$(curl -s -F \"reqtype=fileupload\" -F \"fileToUpload=@$VIDEO_PATH\" https:\u002F\u002Fcatbox.moe\u002Fuser\u002Fapi.php)\n   ```\n   Do this for both the desktop and mobile videos and keep both URLs.\n\n3. Get the current branch and PR:\n   ```bash\n   PR_JSON=$(gh pr view --json number,url 2>\u002Fdev\u002Fnull)\n   ```\n\n   If no PR exists, create one:\n   ```bash\n   BRANCH=$(git branch --show-current)\n   gh pr create --title \"$BRANCH\" --body \"## Summary\n\n   (auto-created by video-test skill)\n\n   ? Generated with [Claude Code](https:\u002F\u002Fclaude.com\u002Fclaude-code)\"\n   PR_JSON=$(gh pr view --json number,url)\n   ```\n\n4. Post a PR comment with **both** videos linked:\n   ```bash\n   PR_NUMBER=$(echo \"$PR_JSON\" | jq -r '.number')\n   gh pr comment \"$PR_NUMBER\" --body \"## Video Test\n\n   ▶️ [Desktop recording]($DESKTOP_VIDEO_URL)\n   ▶️ [Mobile recording]($MOBILE_VIDEO_URL)\n\n   Recorded with Playwright on $(date -u +%Y-%m-%dT%H:%M:%SZ)\"\n   ```\n\n5. **Fallback** if catbox upload fails: Tell the user the local path to the .webm file and suggest they drag-and-drop it into the PR on GitHub's web UI. On macOS: `open \u003Cpath>` to preview.\n\n## Step 8: Cleanup\n\n```bash\nscripts\u002Fstop-app.sh $RUN_ID\n```\n\n## Step 9: Report results\n\n- Confirm whether the videos (desktop + mobile) were attached to the PR and the screenshots to the issue\n- Include the PR URL\n- Report the local video\u002Fscreenshot paths as backup\n- **Return the VIDEO_URLs** so callers can attach them to the Multica issue or the PR\n",{"name":5,"description":7},"main","c24d356df2a247ed3d08650881e768f30fdb293d","ui\u002Fsrc\u002Fdemo\u002Faltavox\u002Fskills\u002Fvideo-test","https:\u002F\u002Fgithub.com\u002Fdiegoquinteiro\u002Fharnessfile","https:\u002F\u002Fgithub.com\u002Fdiegoquinteiro","2026-09-28 11:26:35",{"source":8,"skillName":5,"commands":58,"prompt":68,"zipAvailable":69},[59,62,65],{"tool":60,"command":61},"npx","npx skills add diegoquinteiro\u002Fharnessfile --skill video-test",{"tool":63,"command":64},"bunx","bunx skills add diegoquinteiro\u002Fharnessfile --skill video-test",{"tool":66,"command":67},"pnpm","pnpm dlx skills add diegoquinteiro\u002Fharnessfile --skill video-test","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fdiegoquinteiro\u002Fharnessfile\u002Fblob\u002Fmain\u002Fui\u002Fsrc\u002Fdemo\u002Faltavox\u002Fskills\u002Fvideo-test\u002FSKILL.md",true,[],[72,95,118,140,163,176],{"skillNo":73,"name":74,"nameZh":6,"description":75,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":76,"categories":77,"keywords":78,"publishedAt":46},"01M6R6AREK1HJ2354SCCVK0MNS","wireframe","Produce low-fidelity wireframes or storyboards to explore a flow, layout, or idea across multiple options before committing to hi-fi design. Use when the user wants to 'explore options', 'sketch something out', 'see a few directions', or when the problem is open enough that hi-fi work would be wasted.","https:\u002F\u002Fgithub.com\u002Fdiegoquinteiro\u002Fharnessfile\u002Fblob\u002Fmain\u002Fui\u002Fsrc\u002Fdemo\u002Faltavox\u002Fskills\u002Fwireframe\u002FSKILL.md",[],[74,79,80,45,81,82,83,84,41,85,86,87,88,89,90,91,92,93,94,31],"design","produce","open","explore","work","user","sketch","idea","options","across","layout","wireframes","hi","problem","fi","directions",{"skillNo":96,"name":97,"nameZh":6,"description":98,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":99,"categories":100,"keywords":101,"publishedAt":46},"01M6R6ACE612DH2JQCCB6V7QWV","risk-assessment","Assess the delivery risk and ambiguity of an issue. Scores three anchored 1-5 risk dimensions — Code Blast Radius, Complexity, Business Blast Radius — into an overall risk level (Low\u002FMedium\u002FHigh) with safeguards, plus a separate Ambiguity score (how many open questions remain) with the open questions listed. Use this BEFORE work starts, whenever someone wants to know how risky, how big, how complex, or how well-specified an issue is, or asks to \"assess risk\", \"risk assessment\", \"score this issue\", \"how risky is this\", \"what's the blast radius\", \"how complex is this\", \"how ambiguous is this\", or \"how many open questions are there\". Accepts a Multica issue ID, a GitHub issue\u002FPR, or a free-text description as the argument.","https:\u002F\u002Fgithub.com\u002Fdiegoquinteiro\u002Fharnessfile\u002Fblob\u002Fmain\u002Fui\u002Fsrc\u002Fdemo\u002Faltavox\u002Fskills\u002Frisk-assessment\u002FSKILL.md",[],[102,103,104,105,106,107,108,81,109,110,83,111,112,113,114,115,34,116,32,117],"risk assessment","assessment","risk","pr","code","blast radius","radius","how","text","complexity","issue","dimensions","three","delivery","free","how complex",{"skillNo":119,"name":120,"nameZh":6,"description":121,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":122,"categories":123,"keywords":127,"publishedAt":46},"01M6R6A0H1PXVJ5YGAR0GQPFTP","research","Deep parallel codebase research before planning or implementation. Spawns 4 specialized agents that explore behavior, architecture, coding standards, and test coverage simultaneously, then synthesizes findings into a structured RESEARCH.md document. Use this skill when the user says \"research\", \"investigate\", \"explore the codebase\", \"understand the code\", \"deep dive\", or wants thorough analysis of an area before making changes. Also use it as the first step of any complex feature — research first, then plan, then implement.","https:\u002F\u002Fgithub.com\u002Fdiegoquinteiro\u002Fharnessfile\u002Fblob\u002Fmain\u002Fui\u002Fsrc\u002Fdemo\u002Faltavox\u002Fskills\u002Fresearch\u002FSKILL.md",[124],{"categoryNo":125,"categoryName":126,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000021","资料检索与研究",[120,128,26,129,106,130,131,132,82,133,134,135,41,136,137,138,35,34,139,31],"implement","architecture","planning","analysis","investigate","parallel","specialized","coverage","standards","document","deep","deep dive",{"skillNo":141,"name":142,"nameZh":6,"description":143,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":144,"licenseFileUrl":6,"sourceHtmlUrl":145,"categories":146,"keywords":147,"publishedAt":46},"01M6R69XDJ6KYJE6JNEKBP4FAY","ponytail","Lazy senior dev mode for any coding task (write, refactor, fix, review): YAGNI, stdlib first, no unrequested abstractions. Not for non-coding requests.","MIT","https:\u002F\u002Fgithub.com\u002Fdiegoquinteiro\u002Fharnessfile\u002Fblob\u002Fmain\u002Fui\u002Fsrc\u002Fdemo\u002Faltavox\u002Fskills\u002Fponytail\u002FSKILL.md",[],[142,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162],"review","github","dev","https","refactor","first","coding","write","fix","yagni","task","requests","mode","for","lazy",{"skillNo":164,"name":165,"nameZh":6,"description":166,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":144,"licenseFileUrl":6,"sourceHtmlUrl":167,"categories":168,"keywords":169,"publishedAt":46},"01M6R69TX91W2HT26WA7VYHYGP","ponytail-review","Review a diff for over-engineering. Finds what to delete: reinvented stdlib, needless deps, speculative abstractions. One line per finding.","https:\u002F\u002Fgithub.com\u002Fdiegoquinteiro\u002Fharnessfile\u002Fblob\u002Fmain\u002Fui\u002Fsrc\u002Fdemo\u002Faltavox\u002Fskills\u002Fponytail-review\u002FSKILL.md",[],[148,170,142,149,171,151,172,173,174,175,161],"ponytail review","delete","engineering","deps","diff","line",{"skillNo":177,"name":178,"nameZh":6,"description":179,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":180,"categories":181,"keywords":185,"publishedAt":46},"01M6R69PMB2G6QDK39N9VVV6Z3","polish-pass","Run a comprehensive end-of-design quality gate before a design is shown to stakeholders or shipped, covering hierarchy, AI-slop, interaction states, and accessibility. Use as the final review pass before delivery, or when the user asks to 'polish', 'finalize', or 'make it shippable'.","https:\u002F\u002Fgithub.com\u002Fdiegoquinteiro\u002Fharnessfile\u002Fblob\u002Fmain\u002Fui\u002Fsrc\u002Fdemo\u002Faltavox\u002Fskills\u002Fpolish-pass\u002FSKILL.md",[182],{"categoryNo":183,"categoryName":184,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000015","视觉设计",[186,187,79,148,188,189,190,191,41,192,193,115,194,34,195,196,197,198,199,31],"polish","polish pass","accessibility","quality gate","run","quality","gate","design is","end","interaction","it","states","slop","hierarchy"]