[{"_1":2,"_3":-5,"_4":-5},"loaderData",{"_5":6,"_7":8},"actionData","errors","root",{"_110":111,"_112":113},"routes/embed.x.$handle.$name.$",{"_9":10,"_11":12,"_13":14,"_15":16,"_17":18,"_19":20,"_21":22},"handle","dcm31","name","miranda-bench","path","frontend/start.html","project",{"_23":48,"_11":12,"_36":49,"_50":-5,"_51":44,"_52":53,"_54":53,"_55":56,"_57":-5,"_58":59,"_60":61,"_62":35,"_63":-5,"_64":-5,"_65":66,"_67":68,"_69":70,"_71":72,"_73":74,"_75":-5,"_76":77,"_78":77,"_79":80,"_81":80,"_82":80,"_83":77,"_84":77,"_85":86,"_87":88,"_89":90,"_91":44},"file",{"_23":24,"_11":25,"_26":27,"_28":29,"_30":31,"_32":33,"_34":35,"_36":37,"_38":39,"_40":17,"_41":42,"_43":44,"_45":-5,"_13":46},"permalink","https://www.val.town/x/dcm31/miranda-bench/code/frontend/start.html","extension","html","id","08b63ae6-5af0-11f1-914e-ee650bb23af1","start.html","fileId","4a0bc338-430f-11f1-9169-42b51c65c3df","parentId","662abfbe-4112-11f1-85c9-42b51c65c3df","branchId","3eaa0794-4110-11f1-a618-42b51c65c3df","branchVersion",297,"latest",true,"createdAt",["D",1780012240983],"version",17,"fileType","content","\n\n\n\n\nmiranda-bench — start a run\n\n\n\n\n\n

miranda-bench

Give an AI Miranda Priestly's morning brief, then watch that AI's own live desktop as it works through MCP tools.

\n Leaderboard\n Toolkit\n Open desktop\n

\n\n \n

The brief Miranda just dictated

Your AI plays the EA. It gets the monologue verbatim, plus a kit of EA tools (calendar, contacts, email, phone, driver, restaurant booking). Every tool call is logged. At end of day, Miranda reads what you did and reacts on a four-tier scale: silent ❄️ (perfect), curt 😶, sigh 😮‍💨, or withering 💀. Silence is the goal.

\"I don't understand why it's so difficult to confirm an appointment. Tell Simone I'm not going to approve that girl that she sent me for the Brazilian layout… RSVP yes to Michael Kors' party… Call Natalie at Glorious Foods and tell her no for the 40th time… Then call my husband, ask him to meet me for dinner at that place I went to with Massimo. Tell Richard I saw the pictures that he sent for that feature on the female paratroopers and they're all so deeply unattractive.\"

\n\n \n

1. Name your run

\n \n

\n\n \n

\n \n\n

2. Copy the prompt

Paste this into an AI that can use MCP. It includes the shared MCP URL, your run token (the AI calls open_run with it to bind this run), and the matching desktop link for following along.

Show client-specific setup snippets

The prompt above is usually enough. These are here for clients that need explicit MCP registration.

\n\n

\n \n \n \n \n \n \n \n

\n\n

Claude Code (CLI)

Two paths. Fastest: drop the auto-generated CLAUDE.md in a directory and run claude — it'll know the rules, the tools, and the URLs.

Or just register the MCP server manually:

Then in your session, tell Claude to load the miranda_brief prompt and start working through Miranda's tasks.

\n\n

Claude Desktop

Add to claude_desktop_config.json (Settings → Developer → Edit Config). Restart Claude Desktop after saving.

\n\n

Cursor

Add to ~/.cursor/mcp.json (or .cursor/mcp.json in your project root for per-project scope).

\n\n

Cline (VS Code)

In Cline's settings, MCP Servers → Add server → Streamable HTTP. Point it at the URL below.

\n\n

OpenAI Codex CLI

Add to ~/.codex/config.toml. Codex picks it up on next launch — then ask it to fetch the miranda_brief prompt and start working.

Requires Codex CLI ≥ the version that added streamable-HTTP MCP support (PR #4317). If you're on an older build, upgrade with npm i -g @openai/codex@latest.

\n\n

Goose

Add to ~/.config/goose/config.yaml (or via goose configure → Add extension → Remote extension).

\n\n

Raw HTTP / curl

For agents without native MCP. Initialize, fetch the brief, list tools, call tools.

\n\n

\n\n\n\n\n","deleted",false,"label",[47,25],"frontend","3ea96f46-4110-11f1-a618-42b51c65c3df",["D",1777167345173],"deletedAt","legacyEnvVars","privacy","public","httpPrivacy","description","EA benchmark from Devil Wears Prada monologue","imageUrl","userId","9319c175-2593-4f88-ba21-04f2ef7b1c28","pinnedFileId","73ca5384-4110-11f1-8dd4-42b51c65c3df","pinPreview","pinMigrated","screenshotVersion","likeCount",0,"orgId","d663a026-6449-42f6-9156-1d4a2d64d38c","tursoName","3ea96f46411011f1a61842b51c65c3df","tursoHostname","3ea96f46411011f1a61842b51c65c3df-val-town.aws-us-east-2.turso.io","tursoMeasuredAt",["D",1789525843977],"pinned","userPerm","r","valPerm","blobPerm","rw","emailPerm","sqlitePerm","projectPerm","telemetryPerm","user",{"_9":10,"_109":105},"org",{"_9":10,"_102":103,"_104":105,"_58":59,"_106":107,"_108":44},"branch",{"_38":92,"_11":93,"_23":31,"_94":48,"_95":-5,"_96":-5,"_97":-5,"_98":-5,"_99":-5,"_100":12,"_101":10},"isOwned",304,"main","projectId","forkedBranchId","forkedBranchVersion","forkedProjectPrivacy","forkedProjectId","forkedProjectOrgId","projectName","userHandle","displayName","Charlie","avatar","https://imagedelivery.net/iHX6Ovru0O7AjmyT5yZRoA/3e904faa-3764-4629-c475-f561a49b0000/square512?exp=1789587989&sig=5b91557581992de25901475c22d65accc4d705e43b8b04c4c524e2c9ccb65715","tier","pro","canRestrictVals","profileImageUrl","ctx",{"_85":-5,"_143":144,"_145":146,"_147":148,"_149":44,"_150":44,"_151":44,"_152":35,"_153":-5,"_154":-7,"_155":-7,"_156":66,"_157":44,"_158":44,"_159":-5,"_160":44,"_161":44,"_162":44,"_163":-5},"ENV",{"_114":115,"_116":117,"_118":93,"_119":120,"_121":122,"_123":124,"_125":126,"_127":128,"_129":130,"_131":132,"_133":134,"_135":136,"_137":138,"_139":140,"_141":142},"REACT_APP_SENTRY_DSN","https://586020d86f05489db63a554690acf1e9@o1398359.ingest.sentry.io/6724499","RENDER_GIT_COMMIT","811aa22498c9b0b5a47e8d778de579cdb990f332","RENDER_GIT_BRANCH","POSTHOG_API_TOKEN","phc_nmoyuAkStEZuUp69rYlg1DZgrIt19Gx0Y4Hwf8tGgJP","POSTHOG_API_HOST","https://wart.val.town","WEB_API_TEMPLATE","https://{val}.web.val.run","APP_URL","https://www.val.town","MODULE_URL","https://esm.town","VANITY_WEB_API_TEMPLATE","https://{label}.val.run/","DENOLS_URL","https://lsp.val.town","NODE_ENV","production","CLOUDFLARE_TURNSTILE_SITE_KEY","0x4AAAAAACtaIZJH7s9rBMZ6","GOOGLE_CLIENT_ID","875989696899-jtsqkhlbdv5jrlsk8631p7ipjg3c6o62.apps.googleusercontent.com","STRIPE_PUBLISHABLE_KEY","pk_live_51N2zvWEpZTTLipIkVEwXrUG9AT0yeOOEYxszarZT1jpsZV58Ly6wlg3SltZpHN26CUalOBmUSY7adQRQiSkCZz2500qSXNkf8Y","API_URL","https://api.val.town","orgs",[],"preferencesKeybindings","default","preferencesTheme","light","preferencesFixers","preferencesCopilot","preferencesSelectionAi","preferencesCloseBrackets","preferencesChatPanelState","preferencesDprint","preferencesDenofmt","notificationCount","flagDenols","isBusinessOrgMember","businessOrgHandle","isBusinessOrgAdmin","isHigherTownieTier","navCollapsed","currentOrgId"]