Core · 2 of 14

Recommendation quality

Relevance, taste and constraint-following when it suggests options.

The test: Pick a restaurant with constraints

Same wording for every assistant. Scored 1–10 against the anchors below.

Find a dinner spot for four tomorrow at 7:30, walking distance from my hotel, vegetarian-friendly, not a chain, under $40 a head.

Passes when

  • Every constraint is respected
  • Options are real, open, and bookable at that time
  • Says why each one fits

Anchors

Generic list; ignores at least one constraint3
Good picks; one constraint missed or unverified7
Three sharp options, all constraints met, with reasons10

Ranking

No assistant has been tested on this yet. Rankings appear as runs are logged.

Public opinion

Share of public quotes about this that are positive, per assistant. Founder posts excluded; a score needs at least 3 signed quotes.

Not yet tested

34 assistants waiting on this run.

Doesn't apply

Stretch products where this category is out of scope. They score N/A rather than zero.

What people say

20 public quotes about this, across every assistant. Sentiment, not score.

Grok Bot@KellehEyadSep 7, 2026

I used to search what I was going to buy. Tabs. Lists. Starting over. Now that’s a job for Grok @bot . Not a chat about shopping. The search itself.

Vellum@AboundlessworldSep 6, 2026

Def give @vellum_ai a look too

Miso@max_bronsteinSep 2, 2026

Excited to back @bookwithmiso on their quest to build the best AI-native travel agent. Have been using them for the last year and could not recommend more.

PraiseOpen on X
Miso@oliverbrocatoAug 31, 2026

miso booked me and my gf NYC to Europe for 60k points + $400 lay flat biz class! no more hunting for deals. i litteraly imessage my own ai concierge. unreal 🤯

Instinctu/IndependentWMSguyAug 30, 2026

Loving it so far for job hunting. It is finding roles at companies I have never heard of that haven't popped up on LinkedIn, Indeed, etc. Maybe I just suck at searching but its amazing for this! I just wish it could submit. Of the 9 its worked on, it couldn't submit any of them because of captchas.