[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M79FHVSNBAH3RTKESVX3J5DS":3,"skill-same:01M79FHVSNBAH3RTKESVX3J5DS":59},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":24,"commitSha":25,"skillDirectory":26,"forksCount":14,"repositoryHtmlUrl":27,"repositoryHomepageUrl":6,"ownerHtmlUrl":28,"ownerBlogUrl":6,"ownerDescription":29,"updatedAt":30,"install":31,"resources":44},"01M79FHVSNBAH3RTKESVX3J5DS","replayio",null,"Use when you need to record or inspect an agent browser run in Replay, test a local app with the host agent browser using Replay Chromium, or use the Replay MCP server for deeper debugging of an uploaded recording.","zfifteen\u002Fskills-catalog","zfifteen","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F221906715?v=4","01M79CQFXDZM5J5CJTCECPFA66","Dionisio Alberto Lopez III","01M79CQFY4C62PN8A2T5KDAHGA",1,"MIT","https:\u002F\u002Fgithub.com\u002Fzfifteen\u002Fskills-catalog\u002Fblob\u002Fmain\u002Fskills\u002Freplayio\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: \"replayio\"\ndescription: \"Use when you need to record or inspect an agent browser run in Replay, test a local app with the host agent browser using Replay Chromium, or use the Replay MCP server for deeper debugging of an uploaded recording.\"\n---\n\n# Replay Browser + Agent Browser + Replay MCP\n\nUse the host **agent browser** with Replay Chromium whenever you need a recorded browser session. Do **not** drive the app with `playwright-cli` for normal browser work.\n\nBefore opening the agent browser, point it at Replay Chromium:\n\n```bash\nexport AGENT_BROWSER_EXECUTABLE_PATH=\u002Fpath\u002Fto\u002Fchromium\n```\n\nFor the standard Replay install on macOS, the executable is usually:\n\n```bash\nexport AGENT_BROWSER_EXECUTABLE_PATH=\"$HOME\u002F.replay\u002Fruntimes\u002FReplay-Chromium.app\u002FContents\u002FMacOS\u002FChromium\"\n```\n\nRecordings upload through the plugin's stop\u002Fidle hook as a safety net. If you need the Replay URL before reporting results, close the agent browser tab\u002Fsession and force an upload with `replayio upload-all || replayio upload`.\n\n## Direct Agent Browser First\n\nUse the host agent browser directly for live browser control and first-pass inspection. Do not reach for the Replay MCP server just to click, type, read DOM state, inspect console output, check network requests, take screenshots, read storage, or check cookies.\n\nUse the Replay MCP server only after a recording has uploaded and you need deeper Replay-specific debugging, such as inspecting execution history, narrowing a time-travel debugging problem, or investigating details that the live agent browser cannot answer.\n\nIn Browser-plugin hosts, follow the Browser skill and use the selected `iab` browser. Typical direct checks look like this:\n\n```js\nawait browser.nameSession(\"replay repro\");\nif (typeof tab === \"undefined\") {\n  globalThis.tab = await browser.tabs.new();\n}\nawait tab.goto(URL);\nconsole.log(await tab.playwright.domSnapshot());\nconsole.log(await tab.title());\nconsole.log(await tab.url());\nconsole.log(await tab.dev.logs({ levels: [\"error\"], limit: 50 }));\nawait nodeRepl.emitImage(await tab.screenshot({ fullPage: false }));\n```\n\nUse the browser API's Playwright\u002FDOM\u002Fvision helpers for interaction; the key restriction is to avoid the external `playwright-cli` path.\n\n## Close-When-Done Contract\n\nAfter you finish a test run, **before reporting the outcome to the user**, close the agent browser tab or session using the host browser API.\n\nIn Browser-plugin hosts:\n\n```js\nawait tab.close();\n```\n\nThen, if you need the uploaded Replay URL before your response, run:\n\n```bash\nreplayio upload-all || replayio upload\n```\n\nDo not leave a browser open at the end of your turn. If you forget, the stop\u002Fidle hook will attempt to upload pending recordings as a safety net, but you may not see the resulting Replay URL before responding.\n\n**Exception - authentication wall:** If you must stop because the user needs to sign in interactively (see below), **do not** close the browser just to retry or reset. Leaving the session open preserves the headed window they should use; closing can end the recording before login is done.\n\n## Web App Authentication Walls\n\nIf you hit a **login or authorization barrier** you cannot complete with automation alone - for example a sign-in page, SSO redirect, MFA step, CAPTCHA, or consent screen - **do not** close the browser and loop on reopen\u002Fretry. That drops useful context and trains failing retries.\n\nInstead:\n\n1. **Stop driving the browser** for this turn.\n2. Briefly report what blocked you (URL or visible state).\n3. Ask the user to complete sign-in **in the existing headed browser session** when that is possible (or give them the exact URL if they must use another window).\n4. Ask them to **send another message when they are logged in** so you can attach to the same session or continue from an authenticated page.\n5. End your turn there; resume only after they confirm.\n\nDo not treat an auth wall as a generic error to brute-force by closing and reopening the Replay browser.\n\n## The Reliable Path\n\n1. Verify the Replay runtime exists and the CLI is logged in.\n2. Resolve the Replay Chromium executable path.\n3. Export `AGENT_BROWSER_EXECUTABLE_PATH` to that executable **before** opening the agent browser.\n4. Set both `RECORD_ALL_CONTENT='1'` and `RECORD_REPLAY_VERBOSE='1'`.\n5. If Replay QA will be used, map `REPLAY_QA_API_KEY` from `SECRET_REPLAY_QA_API_KEY` when available.\n6. If testing a local app, start it first and verify the actual reachable URL.\n7. Drive and inspect the page directly with the host agent browser, not `playwright-cli`.\n8. Use fresh DOM snapshots or screenshots after navigation and major UI changes.\n9. Close the agent browser tab\u002Fsession when done.\n10. Run `replayio upload-all || replayio upload` if you need the uploaded Replay URL before reporting results.\n\n## Prerequisites\n\nCheck Replay first:\n\n```bash\nreplayio info\nreplayio whoami\n```\n\nIf Replay is missing, verify `npx` exists:\n\n```bash\ncommand -v npx >\u002Fdev\u002Fnull 2>&1\n```\n\nIf `npx` is missing, stop and ask the user to install Node.js\u002Fnpm. If `npx` exists, install Replay:\n\n```bash\nnpx @replayio\u002Freplay install\n```\n\nIf not logged in, authenticate:\n\n```bash\nreplayio login\n```\n\n## Replay MCP Authentication (Private Recordings)\n\nReplay MCP calls are authorized **per user**. If tools return **Access denied**, you are usually not authenticated to Replay **as the same account that owns the recording**.\n\n- **Cursor**: open **Settings -> MCP**, select the **Replay** server, and complete **Sign in \u002F Connect \u002F Reconnect** so Cursor can finish the **OAuth** flow (including **dynamic client registration** when supported). Stay on `https:\u002F\u002Fdispatch.replay.io\u002Fmcp` (do not swap in unrelated endpoints unless Replay explicitly instructs you to).\n- **Other hosts**: follow that host's MCP authentication mechanism (some environments use API keys or separate app connectors instead of OAuth).\n\n## Agent Browser Executable Path\n\nThe agent browser should launch Replay Chromium through `AGENT_BROWSER_EXECUTABLE_PATH`:\n\n```bash\nexport AGENT_BROWSER_EXECUTABLE_PATH=\"$HOME\u002F.replay\u002Fruntimes\u002FReplay-Chromium.app\u002FContents\u002FMacOS\u002FChromium\"\n```\n\nVerify the executable exists before browser work:\n\n```bash\ntest -x \"$AGENT_BROWSER_EXECUTABLE_PATH\"\n```\n\nDo not configure `.playwright\u002Fcli.config.json` for normal runs, and do not switch back to `playwright-cli` just to select the browser executable. If the agent browser was already running before the environment variable was set, restart or reconnect the agent browser so it picks up the Replay Chromium path.\n\n## Recording Environment\n\nSet recording flags before the run:\n\n```bash\nexport RECORD_ALL_CONTENT='1'\nexport RECORD_REPLAY_VERBOSE='1'\n```\n\nSome hosts or hooks may set these automatically. If in doubt, set them explicitly before opening the agent browser.\n\n## Replay QA Environment\n\nIf Replay QA will be used, map the Replay QA API secret before calling the Replay QA API skill:\n\n```bash\nif [ -z \"${REPLAY_QA_API_KEY:-}\" ] && [ -n \"${SECRET_REPLAY_QA_API_KEY:-}\" ]; then\n  export REPLAY_QA_API_KEY=\"$SECRET_REPLAY_QA_API_KEY\"\nfi\n\ntest -n \"${REPLAY_QA_API_KEY:-}\"\n```\n\nNever print the token. Replay QA tokens start with `lqa_`.\n\n## Local App Check\n\nIf testing a local app:\n\n1. Start the app first.\n2. Use the URL the dev server actually prints.\n3. Do not assume the requested port is the final port. Some dev servers auto-increment when the port is busy.\n4. Verify reachability before opening the browser.\n\n```bash\ncurl -I http:\u002F\u002F127.0.0.1:4323\u002Ftodos\n```\n\nIf a localhost request fails even though a process is clearly listening, you may be in a restricted sandbox. Rerun the browser and reachability checks outside the sandbox.\n\n## Reliable Agent Browser Workflow\n\n```bash\nexport AGENT_BROWSER_EXECUTABLE_PATH=\"$HOME\u002F.replay\u002Fruntimes\u002FReplay-Chromium.app\u002FContents\u002FMacOS\u002FChromium\"\nexport RECORD_ALL_CONTENT='1'\nexport RECORD_REPLAY_VERBOSE='1'\n```\n\nThen use the host agent browser. In Browser-plugin hosts:\n\n```js\nconst URL = \"http:\u002F\u002F127.0.0.1:4323\u002Ftodos\";\nawait browser.nameSession(\"replay todos\");\nif (typeof tab === \"undefined\") {\n  globalThis.tab = await browser.tabs.new();\n}\nawait tab.goto(URL);\nconsole.log(await tab.playwright.domSnapshot());\nawait tab.playwright.getByLabel(\"New todo\", { exact: false }).fill(\"Buy milk\", {});\nawait tab.playwright.getByRole(\"button\", { name: \"Add\" }).click({});\nconsole.log(await tab.playwright.domSnapshot());\nconsole.log(await tab.dev.logs({ levels: [\"error\"], limit: 50 }));\nawait nodeRepl.emitImage(await tab.screenshot({ fullPage: false }));\nawait tab.close();\n```\n\nUse `domSnapshot()` before constructing locators, and again after DOM changes or navigation.\n\n## Attach To An Already-Open Agent Browser\n\nIf the agent browser is already running, attach through the host browser API instead of starting a new CLI session.\n\nIn Browser-plugin hosts:\n\n```js\nconst tabs = await browser.tabs.list();\nconsole.log(tabs);\nglobalThis.tab = tabs.length > 0 ? await browser.tabs.get(tabs[0].id) : await browser.tabs.new();\n```\n\n## Analyzing Uploaded Recordings\n\nFirst inspect the live run with direct agent-browser APIs. Once a recording has uploaded, use the `replay` MCP server tools only when you need deeper Replay-specific debugging beyond direct browser output. Codex connects to Replay through the app configured in `.app.json`, which provides app-level authentication and exposes the Replay MCP tools.\n\n## Replay MCP Widgets\n\nReplay MCP tool calls may return both text `content` for the model and `structuredContent` for an MCP Apps widget. In MCP Apps-aware hosts, prefer the rendered widget for dense debugging views such as Logpoint output, React component trees, Redux actions, network details, screenshots, source code, profiles, and exception stacks.\n\nWhen a widget is visible, use it as evidence instead of restating every detail in prose. Use follow-up actions or related Replay MCP tools when the widget points to a specific point, source, component, request, or stack frame that needs deeper inspection.\n\n## Recording Uploads\n\nThe plugin's stop\u002Fidle hook attempts to upload pending Replay recordings automatically at the end of the turn. Because agent-browser interactions do not necessarily run through a shell close command, force an upload yourself when you need the Replay URL before reporting results:\n\n```bash\nreplayio upload-all || replayio upload\n```\n\nIf you need to inspect the upload state yourself:\n\n```bash\nreplayio list\n```\n\n## Troubleshooting\n\n- If the agent browser does not record, verify `AGENT_BROWSER_EXECUTABLE_PATH` points at Replay Chromium and restart\u002Freconnect the agent browser after setting it.\n- If `test -x \"$AGENT_BROWSER_EXECUTABLE_PATH\"` fails, run `npx @replayio\u002Freplay install` or fix the path.\n- If no Replay URL is available before you respond, close the agent browser tab\u002Fsession and run `replayio upload-all || replayio upload`.\n- If the app is on localhost, verify the exact URL with `curl -I` before opening the browser.\n- If the requested port was busy, use the actual port printed by the dev server.\n- Prefer direct agent-browser inspection (DOM snapshots, console logs, screenshots, storage, cookies, network tools when available) before using the Replay MCP server.\n- If Replay authentication fails, run `replayio login` or reconnect the relevant Replay app\u002Fintegration.\n- If Replay QA returns 401, verify `REPLAY_QA_API_KEY` is set from `SECRET_REPLAY_QA_API_KEY` and starts with `lqa_`.\n- If the **application under test** requires interactive login, follow **Web App Authentication Walls** - do not close-and-retry the browser session in a loop.\n\n## References\n\n- Agent browser reference: `references\u002Fcli.md`\n- Workflow notes: `references\u002Fworkflows.md`\n- [Replay docs](https:\u002F\u002Fdocs.replay.io)\n","# Replay Browser + Agent Browser + Replay MCP\n\nUse the host **agent browser** with Replay Chromium whenever you need a recorded browser session. Do **not** drive the app with `playwright-cli` for normal browser work.\n\nBefore opening the agent browser, point it at Replay Chromium:\n\n```bash\nexport AGENT_BROWSER_EXECUTABLE_PATH=\u002Fpath\u002Fto\u002Fchromium\n```\n\nFor the standard Replay install on macOS, the executable is usually:\n\n```bash\nexport AGENT_BROWSER_EXECUTABLE_PATH=\"$HOME\u002F.replay\u002Fruntimes\u002FReplay-Chromium.app\u002FContents\u002FMacOS\u002FChromium\"\n```\n\nRecordings upload through the plugin's stop\u002Fidle hook as a safety net. If you need the Replay URL before reporting results, close the agent browser tab\u002Fsession and force an upload with `replayio upload-all || replayio upload`.\n\n## Direct Agent Browser First\n\nUse the host agent browser directly for live browser control and first-pass inspection. Do not reach for the Replay MCP server just to click, type, read DOM state, inspect console output, check network requests, take screenshots, read storage, or check cookies.\n\nUse the Replay MCP server only after a recording has uploaded and you need deeper Replay-specific debugging, such as inspecting execution history, narrowing a time-travel debugging problem, or investigating details that the live agent browser cannot answer.\n\nIn Browser-plugin hosts, follow the Browser skill and use the selected `iab` browser. Typical direct checks look like this:\n\n```js\nawait browser.nameSession(\"replay repro\");\nif (typeof tab === \"undefined\") {\n  globalThis.tab = await browser.tabs.new();\n}\nawait tab.goto(URL);\nconsole.log(await tab.playwright.domSnapshot());\nconsole.log(await tab.title());\nconsole.log(await tab.url());\nconsole.log(await tab.dev.logs({ levels: [\"error\"], limit: 50 }));\nawait nodeRepl.emitImage(await tab.screenshot({ fullPage: false }));\n```\n\nUse the browser API's Playwright\u002FDOM\u002Fvision helpers for interaction; the key restriction is to avoid the external `playwright-cli` path.\n\n## Close-When-Done Contract\n\nAfter you finish a test run, **before reporting the outcome to the user**, close the agent browser tab or session using the host browser API.\n\nIn Browser-plugin hosts:\n\n```js\nawait tab.close();\n```\n\nThen, if you need the uploaded Replay URL before your response, run:\n\n```bash\nreplayio upload-all || replayio upload\n```\n\nDo not leave a browser open at the end of your turn. If you forget, the stop\u002Fidle hook will attempt to upload pending recordings as a safety net, but you may not see the resulting Replay URL before responding.\n\n**Exception - authentication wall:** If you must stop because the user needs to sign in interactively (see below), **do not** close the browser just to retry or reset. Leaving the session open preserves the headed window they should use; closing can end the recording before login is done.\n\n## Web App Authentication Walls\n\nIf you hit a **login or authorization barrier** you cannot complete with automation alone - for example a sign-in page, SSO redirect, MFA step, CAPTCHA, or consent screen - **do not** close the browser and loop on reopen\u002Fretry. That drops useful context and trains failing retries.\n\nInstead:\n\n1. **Stop driving the browser** for this turn.\n2. Briefly report what blocked you (URL or visible state).\n3. Ask the user to complete sign-in **in the existing headed browser session** when that is possible (or give them the exact URL if they must use another window).\n4. Ask them to **send another message when they are logged in** so you can attach to the same session or continue from an authenticated page.\n5. End your turn there; resume only after they confirm.\n\nDo not treat an auth wall as a generic error to brute-force by closing and reopening the Replay browser.\n\n## The Reliable Path\n\n1. Verify the Replay runtime exists and the CLI is logged in.\n2. Resolve the Replay Chromium executable path.\n3. Export `AGENT_BROWSER_EXECUTABLE_PATH` to that executable **before** opening the agent browser.\n4. Set both `RECORD_ALL_CONTENT='1'` and `RECORD_REPLAY_VERBOSE='1'`.\n5. If Replay QA will be used, map `REPLAY_QA_API_KEY` from `SECRET_REPLAY_QA_API_KEY` when available.\n6. If testing a local app, start it first and verify the actual reachable URL.\n7. Drive and inspect the page directly with the host agent browser, not `playwright-cli`.\n8. Use fresh DOM snapshots or screenshots after navigation and major UI changes.\n9. Close the agent browser tab\u002Fsession when done.\n10. Run `replayio upload-all || replayio upload` if you need the uploaded Replay URL before reporting results.\n\n## Prerequisites\n\nCheck Replay first:\n\n```bash\nreplayio info\nreplayio whoami\n```\n\nIf Replay is missing, verify `npx` exists:\n\n```bash\ncommand -v npx >\u002Fdev\u002Fnull 2>&1\n```\n\nIf `npx` is missing, stop and ask the user to install Node.js\u002Fnpm. If `npx` exists, install Replay:\n\n```bash\nnpx @replayio\u002Freplay install\n```\n\nIf not logged in, authenticate:\n\n```bash\nreplayio login\n```\n\n## Replay MCP Authentication (Private Recordings)\n\nReplay MCP calls are authorized **per user**. If tools return **Access denied**, you are usually not authenticated to Replay **as the same account that owns the recording**.\n\n- **Cursor**: open **Settings -> MCP**, select the **Replay** server, and complete **Sign in \u002F Connect \u002F Reconnect** so Cursor can finish the **OAuth** flow (including **dynamic client registration** when supported). Stay on `https:\u002F\u002Fdispatch.replay.io\u002Fmcp` (do not swap in unrelated endpoints unless Replay explicitly instructs you to).\n- **Other hosts**: follow that host's MCP authentication mechanism (some environments use API keys or separate app connectors instead of OAuth).\n\n## Agent Browser Executable Path\n\nThe agent browser should launch Replay Chromium through `AGENT_BROWSER_EXECUTABLE_PATH`:\n\n```bash\nexport AGENT_BROWSER_EXECUTABLE_PATH=\"$HOME\u002F.replay\u002Fruntimes\u002FReplay-Chromium.app\u002FContents\u002FMacOS\u002FChromium\"\n```\n\nVerify the executable exists before browser work:\n\n```bash\ntest -x \"$AGENT_BROWSER_EXECUTABLE_PATH\"\n```\n\nDo not configure `.playwright\u002Fcli.config.json` for normal runs, and do not switch back to `playwright-cli` just to select the browser executable. If the agent browser was already running before the environment variable was set, restart or reconnect the agent browser so it picks up the Replay Chromium path.\n\n## Recording Environment\n\nSet recording flags before the run:\n\n```bash\nexport RECORD_ALL_CONTENT='1'\nexport RECORD_REPLAY_VERBOSE='1'\n```\n\nSome hosts or hooks may set these automatically. If in doubt, set them explicitly before opening the agent browser.\n\n## Replay QA Environment\n\nIf Replay QA will be used, map the Replay QA API secret before calling the Replay QA API skill:\n\n```bash\nif [ -z \"${REPLAY_QA_API_KEY:-}\" ] && [ -n \"${SECRET_REPLAY_QA_API_KEY:-}\" ]; then\n  export REPLAY_QA_API_KEY=\"$SECRET_REPLAY_QA_API_KEY\"\nfi\n\ntest -n \"${REPLAY_QA_API_KEY:-}\"\n```\n\nNever print the token. Replay QA tokens start with `lqa_`.\n\n## Local App Check\n\nIf testing a local app:\n\n1. Start the app first.\n2. Use the URL the dev server actually prints.\n3. Do not assume the requested port is the final port. Some dev servers auto-increment when the port is busy.\n4. Verify reachability before opening the browser.\n\n```bash\ncurl -I http:\u002F\u002F127.0.0.1:4323\u002Ftodos\n```\n\nIf a localhost request fails even though a process is clearly listening, you may be in a restricted sandbox. Rerun the browser and reachability checks outside the sandbox.\n\n## Reliable Agent Browser Workflow\n\n```bash\nexport AGENT_BROWSER_EXECUTABLE_PATH=\"$HOME\u002F.replay\u002Fruntimes\u002FReplay-Chromium.app\u002FContents\u002FMacOS\u002FChromium\"\nexport RECORD_ALL_CONTENT='1'\nexport RECORD_REPLAY_VERBOSE='1'\n```\n\nThen use the host agent browser. In Browser-plugin hosts:\n\n```js\nconst URL = \"http:\u002F\u002F127.0.0.1:4323\u002Ftodos\";\nawait browser.nameSession(\"replay todos\");\nif (typeof tab === \"undefined\") {\n  globalThis.tab = await browser.tabs.new();\n}\nawait tab.goto(URL);\nconsole.log(await tab.playwright.domSnapshot());\nawait tab.playwright.getByLabel(\"New todo\", { exact: false }).fill(\"Buy milk\", {});\nawait tab.playwright.getByRole(\"button\", { name: \"Add\" }).click({});\nconsole.log(await tab.playwright.domSnapshot());\nconsole.log(await tab.dev.logs({ levels: [\"error\"], limit: 50 }));\nawait nodeRepl.emitImage(await tab.screenshot({ fullPage: false }));\nawait tab.close();\n```\n\nUse `domSnapshot()` before constructing locators, and again after DOM changes or navigation.\n\n## Attach To An Already-Open Agent Browser\n\nIf the agent browser is already running, attach through the host browser API instead of starting a new CLI session.\n\nIn Browser-plugin hosts:\n\n```js\nconst tabs = await browser.tabs.list();\nconsole.log(tabs);\nglobalThis.tab = tabs.length > 0 ? await browser.tabs.get(tabs[0].id) : await browser.tabs.new();\n```\n\n## Analyzing Uploaded Recordings\n\nFirst inspect the live run with direct agent-browser APIs. Once a recording has uploaded, use the `replay` MCP server tools only when you need deeper Replay-specific debugging beyond direct browser output. Codex connects to Replay through the app configured in `.app.json`, which provides app-level authentication and exposes the Replay MCP tools.\n\n## Replay MCP Widgets\n\nReplay MCP tool calls may return both text `content` for the model and `structuredContent` for an MCP Apps widget. In MCP Apps-aware hosts, prefer the rendered widget for dense debugging views such as Logpoint output, React component trees, Redux actions, network details, screenshots, source code, profiles, and exception stacks.\n\nWhen a widget is visible, use it as evidence instead of restating every detail in prose. Use follow-up actions or related Replay MCP tools when the widget points to a specific point, source, component, request, or stack frame that needs deeper inspection.\n\n## Recording Uploads\n\nThe plugin's stop\u002Fidle hook attempts to upload pending Replay recordings automatically at the end of the turn. Because agent-browser interactions do not necessarily run through a shell close command, force an upload yourself when you need the Replay URL before reporting results:\n\n```bash\nreplayio upload-all || replayio upload\n```\n\nIf you need to inspect the upload state yourself:\n\n```bash\nreplayio list\n```\n\n## Troubleshooting\n\n- If the agent browser does not record, verify `AGENT_BROWSER_EXECUTABLE_PATH` points at Replay Chromium and restart\u002Freconnect the agent browser after setting it.\n- If `test -x \"$AGENT_BROWSER_EXECUTABLE_PATH\"` fails, run `npx @replayio\u002Freplay install` or fix the path.\n- If no Replay URL is available before you respond, close the agent browser tab\u002Fsession and run `replayio upload-all || replayio upload`.\n- If the app is on localhost, verify the exact URL with `curl -I` before opening the browser.\n- If the requested port was busy, use the actual port printed by the dev server.\n- Prefer direct agent-browser inspection (DOM snapshots, console logs, screenshots, storage, cookies, network tools when available) before using the Replay MCP server.\n- If Replay authentication fails, run `replayio login` or reconnect the relevant Replay app\u002Fintegration.\n- If Replay QA returns 401, verify `REPLAY_QA_API_KEY` is set from `SECRET_REPLAY_QA_API_KEY` and starts with `lqa_`.\n- If the **application under test** requires interactive login, follow **Web App Authentication Walls** - do not close-and-retry the browser session in a loop.\n\n## References\n\n- Agent browser reference: `references\u002Fcli.md`\n- Workflow notes: `references\u002Fworkflows.md`\n- [Replay docs](https:\u002F\u002Fdocs.replay.io)\n",{"name":5,"description":7},"main","d9535210da9af28aa76bca80de80355197e1b6d6","skills\u002Freplayio","https:\u002F\u002Fgithub.com\u002Fzfifteen\u002Fskills-catalog","https:\u002F\u002Fgithub.com\u002Fzfifteen","I like to measure things.","2026-07-19 23:41:54",{"source":8,"skillName":5,"commands":32,"prompt":42,"zipAvailable":43},[33,36,39],{"tool":34,"command":35},"npx","npx skills add zfifteen\u002Fskills-catalog --skill replayio",{"tool":37,"command":38},"bunx","bunx skills add zfifteen\u002Fskills-catalog --skill replayio",{"tool":40,"command":41},"pnpm","pnpm dlx skills add zfifteen\u002Fskills-catalog --skill replayio","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fzfifteen\u002Fskills-catalog\u002Fblob\u002Fmain\u002Fskills\u002Freplayio\u002FSKILL.md",true,[45,51,55],{"path":46,"resourceType":47,"resourceTypeDesc":48,"sizeBytes":49,"sourceHtmlUrl":50},"agents\u002Fopenai.yaml","REFERENCE","参考文档",199,"https:\u002F\u002Fgithub.com\u002Fzfifteen\u002Fskills-catalog\u002Fblob\u002Fmain\u002Fskills\u002Freplayio\u002Fagents\u002Fopenai.yaml",{"path":52,"resourceType":47,"resourceTypeDesc":48,"sizeBytes":53,"sourceHtmlUrl":54},"references\u002Fcli.md",1398,"https:\u002F\u002Fgithub.com\u002Fzfifteen\u002Fskills-catalog\u002Fblob\u002Fmain\u002Fskills\u002Freplayio\u002Freferences\u002Fcli.md",{"path":56,"resourceType":47,"resourceTypeDesc":48,"sizeBytes":57,"sourceHtmlUrl":58},"references\u002Fworkflows.md",1335,"https:\u002F\u002Fgithub.com\u002Fzfifteen\u002Fskills-catalog\u002Fblob\u002Fmain\u002Fskills\u002Freplayio\u002Freferences\u002Fworkflows.md",[60,67,74,81,90,97],{"skillNo":61,"name":62,"nameZh":6,"description":63,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":64,"categories":65,"keywords":66,"publishedAt":19},"01M79GXQ2PWPT2E4CJYHXVJ0YE","Zotero","Use Zotero Desktop from Codex to enable\u002Fprobe the local API, search a local Zotero library, list items\u002Fcollections\u002Ftags, export BibTeX, insert citation keys into LaTeX or Markdown drafts, read indexed full text when requested, and import BibTeX\u002FRIS records into Zotero through the connector server. Use when the user mentions Zotero, citations, references.bib, BibTeX export, local Zotero API, localhost:23119, or adding citations from a Zotero library.","https:\u002F\u002Fgithub.com\u002Fzfifteen\u002Fskills-catalog\u002Fblob\u002Fmain\u002Fskills\u002Fzotero\u002FSKILL.md",[],[],{"skillNo":68,"name":69,"nameZh":6,"description":70,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":71,"categories":72,"keywords":73,"publishedAt":19},"01M79GXN85NETKBE06865CWK6Z","zoom-apps-sdk","Use when using Apps SDK.","https:\u002F\u002Fgithub.com\u002Fzfifteen\u002Fskills-catalog\u002Fblob\u002Fmain\u002Fskills\u002Fzoom-apps-sdk\u002FSKILL.md",[],[],{"skillNo":75,"name":76,"nameZh":6,"description":77,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":78,"categories":79,"keywords":80,"publishedAt":19},"01M79GXCYD7FGYS932J9N9RFFY","youtube","Obtain accurate, timestamped YouTube video transcripts (preferring manual captions over auto-generated) as the primary deliverable, then derive summaries, key claims, or notes from that transcript.  Use when the user provides a YouTube URL (or video ID) and wants the transcript, a faithful summary, key claims with attribution, timestamped notes, or \"what does this video actually say\".  The skill is transcript-first: never substitute metadata, title, description, or search snippets for the spoken content.  Triggers: \"youtube transcript\", \"get the transcript of this video\", \"summarize this youtube video\", \"what did they say in https:\u002F\u002Fyoutube.com\u002F...\", \"key claims from this talk\".","https:\u002F\u002Fgithub.com\u002Fzfifteen\u002Fskills-catalog\u002Fblob\u002Fmain\u002Fskills\u002Fyoutube\u002FSKILL.md",[],[],{"skillNo":82,"name":83,"nameZh":6,"description":84,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":85,"licenseFileUrl":86,"sourceHtmlUrl":87,"categories":88,"keywords":89,"publishedAt":19},"01M79GX8EXHHYM566BKR7MS778","yeet","Publish local changes to GitHub by confirming scope, committing intentionally, pushing the branch, and opening a draft PR through the GitHub app from this plugin, with `gh` used only as a fallback where connector coverage is insufficient.","Apache-2.0","https:\u002F\u002Fgithub.com\u002Fzfifteen\u002Fskills-catalog\u002Fblob\u002Fmain\u002Fskills\u002Fyeet\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Fzfifteen\u002Fskills-catalog\u002Fblob\u002Fmain\u002Fskills\u002Fyeet\u002FSKILL.md",[],[],{"skillNo":91,"name":92,"nameZh":6,"description":93,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":94,"categories":95,"keywords":96,"publishedAt":19},"01M79GX2HFJPX84NMRE5VBTYX9","x-post-creator","Create X-ready post threads and co-located story packages from the active conversation or referenced project material. Use when the user says turn this into an X-post, \u002Fx-post, make this into an X thread, or wants plain-language social posts with supporting images saved to docs\u002Fx-posts\u002F[story-name]\u002F.","https:\u002F\u002Fgithub.com\u002Fzfifteen\u002Fskills-catalog\u002Fblob\u002Fmain\u002Fskills\u002Fx-post-creator\u002FSKILL.md",[],[],{"skillNo":98,"name":99,"nameZh":6,"description":100,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":101,"categories":102,"keywords":103,"publishedAt":19},"01M79GWXQT4HBT6QME6QSBZJFA","x-api","Search, read, and write X posts through the local X API MCP server (the same backend used by the Codex x-api skill). Use for searching X\u002Fx.com, reading threads and replies, creating lists, staging\u002Fpublishing posts for @alltheputs, finding high-visibility reply targets for research outreach, ingesting public context, or any request that mentions \u002Fx-api or \"search X\".","https:\u002F\u002Fgithub.com\u002Fzfifteen\u002Fskills-catalog\u002Fblob\u002Fmain\u002Fskills\u002Fx-api\u002FSKILL.md",[],[]]