# Claude Duet (Anthropic)

Feedback Bench, personal agents, built 2026-11-09, window 2026-08-31 to 2026-11-08. Web page: https://feedbackbench.com/#/agent/duet

| Measure | Value |
|---|---|
| Rank | 7 of 13 (rank range 6–8) |
| Feedback Score | 47.9 (95% interval 46.3–49.4) |
| Popularity | 0.471 (share of voice 1.87%) |
| Customer love | 0.489 (95% interval 0.475–0.502) |
| Place | Popularity 7th of 13 × Customer love 8th of 13 |
| Top quadrant | no |
| Users | 2295 |
| Posts counted | 2729 |
| Posts that judge the agent | 330 |
| Criteria ranking high / low | 0 / 0 of 53 |

## The brief

Written by large language models from this page's numbers and the posts it quotes, then checked: every number in the text is in the build. Interpretation, not measurement.

Users want Claude Duet to reach the rest of their Claude work and their accounts. Posts describe it stopping when it cannot. Users praise how it gets tasks done and works on its own, and in both areas praise outnumbers complaints. Connections pull it down. Connecting apps and devices is 16% of what its users rate, 2.1 times the rate for other agents. It is the only area rated worse than the category norm. Users also say Duet uses up their shared Claude usage limits. It ranks 7th of 13 overall. Some posts name Grok Bot and Muse as agents that got past bot checks where Duet stopped.

### Why people choose it


### Why people leave

- **It cannot reach your other Claude work or accounts** ([Connecting apps and devices](https://feedbackbench.com/criteria/connect.md)). 2.1×: 16% of what users rate, 2.1× the rate for other agents. Ranks 5th of 8 on “Connecting apps and devices”. Users say Duet cannot see their Claude Code sessions, their existing chats or their own computer. Posts describe it stopping at captchas and bot checks, and failing to sign into accounts in its own browser. Some users say they went back to other agents for browser logins and API keys.
  Evidence:
  - 2026-10-31, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “finally trying claude duet and kinda bummed it can’t access any of my existing claude chats and can’t use my computer? i have so much context in there” [source](https://twitter.com/607389197/status/2116550154719211733)
  - 2026-10-21, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “claude duet seems to have trouble signing into accounts in its own browser. it can't use any access tokens either, like claude scheduled task can, even when they run in the cloud.” [source](https://twitter.com/2090061596059906049/status/2113057162977272197)
- **It uses up shared usage limits** ([Price and value](https://feedbackbench.com/criteria/billing.md)). 6th: of 9 on “Price and value”. 15% of what users rate. Users say Duet draws on the same usage pool as their other Claude plans. Posts say it keeps spending that usage in the background, on morning reports and even when idle. Some describe hitting limits soon after they got access, and losing a banked reset after a billing change.
  Evidence:
  - 2026-10-23, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “good to know about claude duet. if you don’t check these boxes, duet will burn usage every few hours going over them. burned $60 overnight on idle. i know it has to be proactive, but wow that was unexpected. [link]” [source](https://twitter.com/1605622754430025729/status/2113701125149139097)
  - 2026-10-27, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “unfortunately, i have stopped using claude duet when i woke up to my 5-hour limit 13% used when claude duet had only asked what it should help me with.” [source](https://twitter.com/2090061596059906049/status/2115066595428999199)

### What would move its score

Rule: For each area where an agent's customer love is below the category norm: the Feedback Score and rank it would have if that one area were at the norm (0.5) and everything else stayed the same, computed with the same formula as the ranking. An area's pull follows its weight, so an area many users of the category rate moves the score more than one they rarely mention.

| Area at the category norm | Feedback Score | Change | Position |
|---|---|---|---|
| Getting tasks done | 48.6 | +0.7 | 6th of 13 |
| Running on its own | 48.2 | +0.3 | 6th of 13 |
| Connecting accounts, sites and devices | 48.1 | +0.3 | 6th of 13 |
| Talking to the agent | 48.0 | +0.1 | 6th of 13 |
| Getting access and set up | 47.9 | +0.1 | 6th of 13 |
| Trusting it with data and safety | 47.9 | +0.0 | 6th of 13 |

No single area moves the score far. Getting tasks done would help most. Bringing it to the category norm would lift the Feedback Score from 47.9 to 48.6 and the rank from 7th to 6th. Doing the same for working on its own or for connecting apps and devices would add 0.3 each.

### Six weeks

No users posted before the week of 19 October, which included the US waitlist research preview on 21 October. That week 1,113 users posted. Users then fell to 826 and 463, and share of voice from 4% to 2%. Rank was 6th both weeks.

## Top requests

What users ask to add or change, most asked first. 173 user-weeks ask for something. Requests do not change the Feedback Score. Rule: A separate pass by a large language model reads every counted post and extracts what the user asks the agent or its vendor to add or change, with the criteria it maps to and a short normalised wording; it does not touch the labels or the Feedback Score. A large language model groups the wordings within each criterion (the first criterion the request maps to) into themes, and a second pass merges themes that ask for the same change under different criteria; a merged theme sits under the criterion most of its requests map to, and code counts it once. A theme counts distinct user-weeks that ask for it, per agent; across agents, one user-week per agent. Themes asked in fewer than 2 user-weeks, and requests that share no theme, are not shown. Examples: up to 3 posts per theme from different users, without slurs: whole posts of 60 to 450 characters with at least 8 words beyond handles and links, posts of at most 280 characters first, then the newest.

| Rank | Request | Criterion | User-weeks | Posts |
|---|---|---|---|---|
| 1 | Access granted to the user's account | [Country availability, waitlists and staged rollout](https://feedbackbench.com/criteria/access.availability.md) | 15 | 15 |
| 2 | Availability in more countries | [Country availability, waitlists and staged rollout](https://feedbackbench.com/criteria/access.availability.md) | 12 | 13 |
| 3 | Control coding agents from personal agent | [Multiple agents, teams and delegation between them](https://feedbackbench.com/criteria/auto.multi_agent.md) | 11 | 11 |
| 4 | Availability in Japan | [Country availability, waitlists and staged rollout](https://feedbackbench.com/criteria/access.availability.md) | 9 | 9 |
| 5 | Faster wider rollout to all users | [Country availability, waitlists and staged rollout](https://feedbackbench.com/criteria/access.availability.md) | 9 | 9 |
| 6 | Early access and beta programs | [Country availability, waitlists and staged rollout](https://feedbackbench.com/criteria/access.availability.md) | 8 | 8 |
| 7 | Availability in the EU and Europe | [Country availability, waitlists and staged rollout](https://feedbackbench.com/criteria/access.availability.md) | 7 | 7 |
| 8 | Faster removal from the waitlist | [Country availability, waitlists and staged rollout](https://feedbackbench.com/criteria/access.availability.md) | 6 | 6 |
| 9 | Interoperability with other agents and tools | [Which services have a connector or integration](https://feedbackbench.com/criteria/connect.coverage.md) | 4 | 4 |
| 10 | Agent built into main chat apps | [Apps offered for phone, tablet and desktop](https://feedbackbench.com/criteria/access.platform.md) | 3 | 3 |
| 11 | Availability in Australia and New Zealand | [Country availability, waitlists and staged rollout](https://feedbackbench.com/criteria/access.availability.md) | 3 | 3 |
| 12 | Computer use on user's own computer | [Access to the user's own phone or computer](https://feedbackbench.com/criteria/connect.local_device.md) | 3 | 3 |

### 1. Access granted to the user's account

- Claude Duet, 2026-11-05, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “dear @claudeai @claudedevs , can you enroll me claude duet, its been a while i am waiting for it.” [source](https://twitter.com/2016340376341155840/status/2118332236588314814)
- Claude Duet, 2026-11-01, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “@darioamodei @anthropicai @claudeai day 7 of tweeting until i get to work on anthropic’s duet ❤️” [source](https://twitter.com/1468498551181303811/status/2116906144845242394)
- Claude Duet, 2026-10-31, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “i need access to claude duet. i'm having so much fun with agents rn.” [source](https://twitter.com/3114630536/status/2116399291455643748)

### 2. Availability in more countries

- Claude Duet, 2026-11-06, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “it’s november ‘26 still don’t have claude duet still don’t have meta muse feels like more and more shit gets released only in the us, with zero eagerness to even bring it abroad really pisses me off” [source](https://twitter.com/1773217467558744064/status/2118569687886024987)
- Claude Duet, 2026-11-05, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “muse isn't available globally dot is too expensive i'll have to wait for claude duet to launch before i can switch 🥲” [source](https://twitter.com/1197533035060748294/status/2118187880980058264)
- Claude Duet, 2026-11-03, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “i need claude duet to be available outside the us so bad [link]” [source](https://twitter.com/1998465745559416832/status/2117718643203317973)

### 3. Control coding agents from personal agent

- Claude Duet, 2026-11-05, App Store (App Store): “swing and a miss… essentially just a standalone chatbot that only has access to existing cowork sessions and code sessions but cannot control them.” [source](https://apps.apple.com/us/app/claude-duet-by-anthropic/id6803276000)
- Claude Duet, 2026-11-04, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “@nvandewetering same prompt to grokbot: [30 seconds later] "done" ... side rant: how is it that grok bot can communicate with claude code and claude duet can't???” [source](https://twitter.com/1855005487051554835/status/2117855450448478413)
- Claude Duet, 2026-11-03, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “if claude duet will connect to my own stations's cc, will be really powerful ever. can't wait to see this feature available soon. [link]” [source](https://twitter.com/1658735839252484096/status/2117412316388217285)

### 4. Availability in Japan

- Claude Duet, 2026-11-08, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “claude duet, when are you coming to japan? i'm waiting with a long neck 😗” [source](https://twitter.com/1981122825978335235/status/2119469068201898293)
- Claude Duet, 2026-10-26, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “@minimalistneko claude duet i forgot! haha i'm looking forward to it! i hope it becomes available in japan soon ^ ^” [source](https://twitter.com/1232270534743691264/status/2114549278521905350)
- Claude Duet, 2026-10-24, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “it's interesting that the ai not only waits for instructions but also organizes work proactively. however, claude duet is currently a research preview for the united states. when it becomes available in japan, i would like to try it for job hunting and schedule management. [link]” [source](https://twitter.com/1765195748763906048/status/2113832831231390038)

### 5. Faster wider rollout to all users

- Claude Duet, 2026-11-05, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “claude duet is looking forward to it...!! please come quickly...!! [link]” [source](https://twitter.com/1163787052145504261/status/2118235516105801842)
- Claude Duet, 2026-10-26, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “@sonderby @andersbaek1 i think they are ahead on the models, but lag behind on many of the new features. projects in desktop is a good example. an extremely underrated feature that deserves much more love. claude duet is still not out, while @bot is already really good.” [source](https://twitter.com/1169712217522081792/status/2114754985812480301)
- Claude Duet, 2026-10-25, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “fable 5.5 is very unlikely to be released this month... anthropic, when i said to take the extra week i didn't mean the whole month lol maybe you guys could make up for it by giving all of us claude duet access? [link]” [source](https://twitter.com/1828475904818921472/status/2114207081402380781)

### 6. Early access and beta programs

- Claude Duet, 2026-10-27, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “@edwinarbus hi edwin, when do i get access for claude duet? please 🙏” [source](https://twitter.com/924470178879913984/status/2115173115592581185)
- Claude Duet, 2026-10-27, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “@claudeai excuse me, ~ahem~, i would like access to claude duet please. i don't like my disagregated workflow using muse with claude code and would prefer to use duet in my workflow as my personal assistant. thank you for your consideration. -devkore” [source](https://twitter.com/62593415/status/2115152211328258339)
- Claude Duet, 2026-10-24, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “@cwrichardkim any chance to get early access to claude duet?” [source](https://twitter.com/2016121883796643840/status/2113939047844634752)

### 7. Availability in the EU and Europe

- Claude Duet, 2026-11-06, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “bring claude duet to the eu, and i’m in. grok bot is available here. what’s holding anthropic back? [link]” [source](https://twitter.com/222760298/status/2118801350322536589)
- Claude Duet, 2026-11-05, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “claude duet, chargpt dots and muse really need to plan a trip to the uk and rest of europe” [source](https://twitter.com/14135486/status/2118352906072006989)
- Claude Duet, 2026-10-25, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “if it weren't for grok bot, we wouldn't even be able to use an agent in europe. no claude duet, no chatgpt dots, no meta muse. the extreme eu regulation is now unsustainable for growth. things must change or the gap with the usa and china will become irretrievable.” [source](https://twitter.com/1823380936639369217/status/2114445430155366466)

### 8. Faster removal from the waitlist

- Claude Duet, 2026-11-01, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “applied for claude duet 5 days ago,still pending. anyone got access yet? put “personal use” as company name — does that affect approval time? 😔 i want to use duet😭😭😭😩” [source](https://twitter.com/2101180046345977856/status/2116914395678953581)
- Claude Duet, 2026-10-27, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “@anthropicai congrats now let me get into claude duet. thanks” [source](https://twitter.com/832487400/status/2115178840259518794)
- Claude Duet, 2026-10-26, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “@claudeai when can we get access to claude duet? rolling out slowly is not very helpful as more and more users are adopting available products from competitors.” [source](https://twitter.com/2111288707424595968/status/2114746009053970599)

### 9. Interoperability with other agents and tools

- Claude Duet, 2026-11-03, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “i really need claude duet to have better integerations w rest of claude. i will forever become a claude customer if we get good integerations in duet. [link]” [source](https://twitter.com/1264189402986602496/status/2117727900371730549)
- Claude Duet, 2026-10-29, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “claude duet first 1 minute impressions: can’t use it at all because i use t3 code and it can’t see my claude threads i’ve been working on. it’s a walled garden inside claude. @bot is better and it took me literally 1 minute to tell. [link]” [source](https://twitter.com/1846213442656784384/status/2115898970396623029)
- Claude Duet, 2026-10-22, Anthropic Just Dropped Claude Duet (Full Breakdown) (YouTube): “i am too, the only thing is i wish there was a really good way to use it with claude code” [source](https://www.youtube.com/watch?v=nv1aaP3dHzs&lc=Ugw1V35PlohNKOPu2Il4AaABAg.AcGWELTeuLpAcHC3aLCgmi)

### 10. Agent built into main chat apps

- Claude Duet, 2026-10-29, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “@miles_brundage could future updates bring claude duet into the main app?” [source](https://twitter.com/1724070962902028288/status/2115842855944245271)

### 11. Availability in Australia and New Zealand

- Claude Duet, 2026-11-04, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “@connortalksai still waiting for claude duet to be released in australia 🇦🇺” [source](https://twitter.com/2094279268343980032/status/2117807947950809181)
- Claude Duet, 2026-10-31, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “hey @claudeai! still eagerly waiting down under in 🇦🇺 to get my hands on new claude projects and claude duet. when can we expect full access in australia? 🦘” [source](https://twitter.com/243241051/status/2116488277419241572)
- Claude Duet, 2026-11-02, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “having seen cuts to grokbot usage today, i’m considering dropping my grok bot sub and upgrading my claude max subscription to ultra if claude duet becomes available in australia. i’d prefer to have all my skills and automations within claude rather than switching between claude code for coding and grokbot for personal agent tasks. #claude #grokbot” [source](https://twitter.com/1270418354558435328/status/2117083985407082994)

### 12. Computer use on user's own computer

- Claude Duet, 2026-11-05, App Store (App Store): “the agent is not helpful. i’m not sure if it doesn’t see the same context on my projects but most of the suggestions and questions are things i have previously tried in my project. it doesn’t connect to my computer where i’m logged into everything and it could do actual work” [source](https://apps.apple.com/us/app/claude-duet-by-anthropic/id6803276000)
- Claude Duet, 2026-10-29, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “after a few days of using @anthropicai claude duet would be the best personal assistant if it could actually interact with your claude artifacts or local machine” [source](https://twitter.com/220209773/status/2115842876265353631)
- Claude Duet, 2026-10-27, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “after using grok bot, dot, and claude duet, i think claude duet has strong potential, only thing i miss the most is that it doesn't have access to my local computer. @ahuja_karan help me out here!” [source](https://twitter.com/1038039838602272768/status/2114914336963543389)

## Ranks highest on

No criterion where it sits clearly in the top half.

## Ranks lowest on

No criterion where it sits clearly in the bottom half.

Position on a criterion: its place among the agents with at least 30 rated user-weeks there, ordered by customer love. It ranks high when both its position and its 95% interval put it in the top half, low when both put it in the bottom half; neither otherwise, or when fewer than 4 agents qualify.

## All 53 criteria

Criterion love: 0.5 is the category norm. n: rated user-weeks.

### Paying for the agent: 6th of 9 (customer love 0.505, n 36)

Users say Duet shares usage with their other Claude plans and spends it in the background, even when idle. Complaints outnumber praise, but love stays within the category norm. Users talk about price 1.29 times as often as for other agents.

| Criterion | Reading | Customer love | 95% interval | n | Praise | Complaint |
|---|---|---|---|---|---|---|
| [Usage allowance size, exhaustion and reset wait](https://feedbackbench.com/criteria/billing.usage_cap.md) | Too few posts | 0.521 | 0.490–0.550 | 29 | 10 | 19 |
| [Agent spends usage on overhead and costly choices](https://feedbackbench.com/criteria/billing.token_efficiency.md) | Too few posts | 0.500 | 0.492–0.512 | 4 | 1 | 3 |
| [Free tier, trial or required plan to start](https://feedbackbench.com/criteria/billing.entry_plan.md) | Too few posts | 0.499 | 0.492–0.506 | 2 | 1 | 1 |
| [Subscription price versus value received](https://feedbackbench.com/criteria/billing.price_level.md) | Too few posts | 0.502 | 0.500–0.508 | 1 | 1 | 0 |
| [Buying, upgrading, cancelling and refunding go wrong](https://feedbackbench.com/criteria/billing.purchase_flow.md) | Too few posts | 0.499 | 0.496–0.500 | 1 | 0 | 1 |
| [Plan contents, bundles and usage meters unclear](https://feedbackbench.com/criteria/billing.plan_clarity.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |

Most recent posts:

- Praise, 2026-11-05, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “@buildwithsid claude duet works really well and doesn’t use too much of my usage” [source](https://twitter.com/2482408872/status/2118325832938627214)
- Praise, 2026-11-04, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “found it! claude duet beta testing has its perks. 3 resets so far [link]” [source](https://twitter.com/2063017791460876288/status/2117848569919226033)
- Praise, 2026-11-02, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “@dogecoinnorway @elonmusk have offloaded 2/3 of my grok bots to claude duet. usage is minimal and better. one man’s opinions.” [source](https://twitter.com/860502235213553664/status/2117104339726823574)
- Complaint, 2026-11-08, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “i like claude duet, however with so much real work to do, i have no usage to spare. especially now that i’m using fable daily.” [source](https://twitter.com/1390030030307303429/status/2119459188266926334)
- Complaint, 2026-11-02, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “so you lose your banked reset (from claude duet) if you downgrade from ultra to max. super cool, thank you so much for doing that @claudedevs” [source](https://twitter.com/1946369271287721984/status/2117341544076161481)
- Complaint, 2026-11-02, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “welp, i just lost my free claude reset due to a billing error. what happened - i got a free reset from claude duet, and a day later my subscription fails to renew; i immediately repurchase my subscription, andddd my reset is completely gone from my account... awesome!” [source](https://twitter.com/1595511689356529664/status/2117356867047543256)

### Getting access and set up: Too few posts (customer love 0.495, n 24)

Users ask for access on their own accounts, a faster rollout, quicker removal from the waitlist and more countries, including Japan and Europe. Complaints outnumber praise, 17 to 7. Love stays within the category norm.

| Criterion | Reading | Customer love | 95% interval | n | Praise | Complaint |
|---|---|---|---|---|---|---|
| [Effort needed to set up and train the agent](https://feedbackbench.com/criteria/access.setup.md) | Too few posts | 0.502 | 0.490–0.514 | 9 | 5 | 4 |
| [Apps offered for phone, tablet and desktop](https://feedbackbench.com/criteria/access.platform.md) | Too few posts | 0.489 | 0.481–0.497 | 7 | 0 | 7 |
| [Country availability, waitlists and staged rollout](https://feedbackbench.com/criteria/access.availability.md) | Too few posts | 0.508 | 0.493–0.528 | 6 | 2 | 4 |
| [Account creation, sign-in and identity checks fail](https://feedbackbench.com/criteria/access.sign_in.md) | Too few posts | 0.498 | 0.494–0.500 | 2 | 0 | 2 |

Most recent posts:

- Praise, 2026-10-29, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “just getting started with claude duet... the onboarding experience has been pretty good so far! i'll be sharing more soon 👀 [link]” [source](https://twitter.com/2967102135/status/2115787670060405222)
- Praise, 2026-10-24, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “lovely simple and sleek onboarding animation in claude duet. i guess since the agent already has so much context about what is going on you can just say a few words about what you need. [link]” [source](https://twitter.com/4363348874/status/2113801901951131750)
- Praise, 2026-10-27, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “this is claude duet muse… felt like a game, childish. didn’t like it. grok bot honestly haven’t played enough with it. but i get confused by like having to make different bots. claude, honestly is so good because i already use claude so much. it’s just natural i don’t have to like educate it too much. they did a nice job integrating it with all my existing claude work. and maybe that’s my issue. w…” [source](https://twitter.com/1084811625851486209/status/2114911530902770042)
- Complaint, 2026-10-27, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “downloaded claude duet. now i have to teach a new agent from scratch how todo everything again.” [source](https://twitter.com/2114786547383902208/status/2114906391496249422)
- Complaint, 2026-10-27, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “@rileybrown i can only get claude duet working on desktop :( how did you get it working on mobile? keeps sending me to 'download' the app (and it's not even offering an update) @rileybrown” [source](https://twitter.com/25053942/status/2115144137481871393)
- Complaint, 2026-10-22, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “claude duet doesn’t have access to the files on my macbook since there is no mac app yet. rough unless testflight has a mac app but i don’t think so” [source](https://twitter.com/2063017791460876288/status/2113129944960450725)

### Connecting accounts, sites and devices: 5th of 8 (customer love 0.465, n 39)

This is the weakest area and the only one rated worse than the category norm. Users say Duet cannot reach their Claude Code sessions, their chats or their own computer, and cannot sign into accounts. Users raise it 2.06 times as often as for other agents.

| Criterion | Reading | Customer love | 95% interval | n | Praise | Complaint |
|---|---|---|---|---|---|---|
| [Which services have a connector or integration](https://feedbackbench.com/criteria/connect.coverage.md) | Too few posts | 0.486 | 0.466–0.505 | 15 | 7 | 8 |
| [Access to the user's own phone or computer](https://feedbackbench.com/criteria/connect.local_device.md) | Too few posts | 0.469 | 0.451–0.486 | 14 | 1 | 13 |
| [Linking accounts and staying logged in](https://feedbackbench.com/criteria/connect.account_link.md) | Too few posts | 0.497 | 0.487–0.510 | 7 | 1 | 6 |
| [Websites block the agent with captchas or IP bans](https://feedbackbench.com/criteria/connect.site_blocking.md) | Too few posts | 0.494 | 0.488–0.499 | 4 | 0 | 4 |
| [Moving files between agent, apps and user](https://feedbackbench.com/criteria/connect.file_transfer.md) | Too few posts | 0.499 | 0.495–0.500 | 1 | 0 | 1 |

Most recent posts:

- Praise, 2026-11-06, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “tested the new claude duet build 🚀 from my phone: - started a new claude code session on my laptop - listed every project on it - sorted them by most recent commit never touched the laptop 😅 [link] [link]” [source](https://twitter.com/4774517295/status/2118658812097839559)
- Praise, 2026-10-29, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “claude duet is pretty crazy so far. had it make a 3d interactive model of my kitchen remodel from only 3 images. it automatically pulled the color of the cabinet knobs from amazon purchase confirmation emails. [link]” [source](https://twitter.com/14296763/status/2115897393409380439)
- Praise, 2026-10-28, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “really enjoying claude duet right now. instinct i loved that it was on text muse has let me connect the most external platforms but i like duet has a ton of channels, and that it’s connected to claude code and cowork which i think puts it over the top for me” [source](https://twitter.com/16689451/status/2115579848479678817)
- Complaint, 2026-11-04, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “@nvandewetering same prompt to grokbot: [30 seconds later] "done" ... side rant: how is it that grok bot can communicate with claude code and claude duet can't???” [source](https://twitter.com/1855005487051554835/status/2117855450448478413)
- Complaint, 2026-11-03, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “@maiab i tried using claude duet to pay some utility bills and its browser was blocked 🤷♀️” [source](https://twitter.com/49407684/status/2117657207026151544)
- Complaint, 2026-11-02, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “claude duet would be much more useful if it actually integrated into my work, which is nowadays in the form of remote control sessions running on my always on macbook sitting in the office. @ahuja_karan please help 🙏” [source](https://twitter.com/44626577/status/2117126159842562143)

### Talking to the agent: 7th of 13 (customer love 0.492, n 47)

Users talk about chatting with Duet 1.26 times as often as for other agents. Praise and complaints are about even, and love stays within the category norm.

| Criterion | Reading | Customer love | 95% interval | n | Praise | Complaint |
|---|---|---|---|---|---|---|
| [Threads, chats and parallel tasks organisation](https://feedbackbench.com/criteria/ask.conversation.md) | Too few posts | 0.511 | 0.495–0.526 | 14 | 9 | 5 |
| [App screens, layout, rendering, localisation and stability](https://feedbackbench.com/criteria/ask.app_client.md) | Too few posts | 0.508 | 0.493–0.524 | 13 | 6 | 7 |
| [Speaking with the agent by voice or call](https://feedbackbench.com/criteria/ask.voice.md) | Too few posts | 0.494 | 0.483–0.506 | 9 | 3 | 6 |
| [Response and task completion speed](https://feedbackbench.com/criteria/ask.speed.md) | Too few posts | 0.498 | 0.487–0.509 | 7 | 3 | 4 |
| [Avatar, personality and tone of replies](https://feedbackbench.com/criteria/ask.persona.md) | Too few posts | 0.489 | 0.465–0.505 | 5 | 3 | 2 |
| [Reaching the agent through messaging apps](https://feedbackbench.com/criteria/ask.channel.md) | Too few posts | 0.505 | 0.500–0.511 | 3 | 3 | 0 |

Most recent posts:

- Praise, 2026-10-29, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “just got claude duet! i love that it has channels! i'm crazy about organizing tasks (while having a general chat like i do with my main grokbot).” [source](https://twitter.com/867618742544551936/status/2115835571910373523)
- Praise, 2026-10-23, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “i've been using claude cowork since it launched in early 2026, used mainly grokbot last month. claude duet has the ui figured out. still some things they need to iron out, but the ui for juggling a bunch of tasks is extremely good. a claude copy of grokbot that just works.” [source](https://twitter.com/2868013240/status/2113629595274539189)
- Praise, 2026-10-21, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “i love the to-do list interface in claude duet! you have to try it [link]” [source](https://twitter.com/901616216929742848/status/2112989580974760092)
- Complaint, 2026-11-04, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “anyone else struggling with the 11min capped voice sessions with claude duet? would love a fix to this as agent voice calls are essential to my workflow.” [source](https://twitter.com/1994283149988384768/status/2117855046276620507)
- Complaint, 2026-10-31, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “i got access to claude duet so far i think the ux is pretty confusing there are so may layers of a chat you can be in its hard to tell where you are and why [link]” [source](https://twitter.com/1855146824/status/2116506221360665011)
- Complaint, 2026-10-29, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “@gabemulley claude duet feedback: if i want to reply to this message, it adds the whole message. this happens with press and hold and swipe right. [link]” [source](https://twitter.com/227366159/status/2115675238189122002)

### Getting tasks done: 7th of 13 (customer love 0.474, n 50)

Users talk about getting tasks done less often than for other agents, 0.69 times the rate. Praise outnumbers complaints about two to one, but love stays within the category norm.

| Criterion | Reading | Customer love | 95% interval | n | Praise | Complaint |
|---|---|---|---|---|---|---|
| [Stops short, or hands the task back](https://feedbackbench.com/criteria/act.stops_short.md) | Too few posts | 0.492 | 0.479–0.504 | 10 | 3 | 7 |
| [Email triage, cleanup and drafting replies](https://feedbackbench.com/criteria/act.email.md) | Too few posts | 0.458 | 0.413–0.498 | 9 | 5 | 4 |
| [Writing code and building apps or websites](https://feedbackbench.com/criteria/act.build.md) | Too few posts | 0.509 | 0.503–0.515 | 8 | 8 | 0 |
| [Creating writing, images, video and social posts](https://feedbackbench.com/criteria/act.content.md) | Too few posts | 0.508 | 0.501–0.515 | 6 | 6 | 0 |
| [Forms, applications, claims and life admin paperwork](https://feedbackbench.com/criteria/act.admin.md) | Too few posts | 0.506 | 0.501–0.511 | 5 | 5 | 0 |
| [Buying products, orders and purchase judgment](https://feedbackbench.com/criteria/act.shopping.md) | Too few posts | 0.505 | 0.500–0.512 | 4 | 4 | 0 |
| [Wrong facts, hallucinations and careless mistakes](https://feedbackbench.com/criteria/act.accuracy.md) | Too few posts | 0.507 | 0.497–0.518 | 4 | 3 | 1 |
| [Calendar entries, conflicts and schedule changes](https://feedbackbench.com/criteria/act.calendar.md) | Too few posts | 0.503 | 0.500–0.507 | 3 | 3 | 0 |
| [Budgets, transactions, bills and investments](https://feedbackbench.com/criteria/act.finance.md) | Too few posts | 0.484 | 0.441–0.506 | 3 | 2 | 1 |
| [Operating a browser or desktop by clicking](https://feedbackbench.com/criteria/act.computer_use.md) | Too few posts | 0.495 | 0.484–0.503 | 3 | 1 | 2 |
| [Contacting other people and businesses for you](https://feedbackbench.com/criteria/act.outreach.md) | Too few posts | 0.501 | 0.500–0.504 | 1 | 1 | 0 |
| [Says it did something it did not do](https://feedbackbench.com/criteria/act.false_reports.md) | Too few posts | 0.499 | 0.497–0.500 | 1 | 0 | 1 |
| [Following instructions, rules and corrections](https://feedbackbench.com/criteria/act.instructions.md) | Too few posts | 0.498 | 0.494–0.500 | 1 | 0 | 1 |
| [Booking appointments, reservations, travel and tickets](https://feedbackbench.com/criteria/act.reservations.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |
| [Web research, monitoring and summarising information](https://feedbackbench.com/criteria/act.research.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |

Most recent posts:

- Praise, 2026-11-06, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “sonnet + notion + agentic ais resulted in a delightful experience this morning: - made a handwritten note with to-dos for today in sonnet - note synced to notion - claude duet picked it up, scanned the handwriting, and added those items to my things to-do list absolute magic” [source](https://twitter.com/1999320417602011137/status/2118780426005459282)
- Praise, 2026-11-02, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “using claude duet to steer and create with claude design is super smooth. results are really really solid.” [source](https://twitter.com/2568921814/status/2117065916869652878)
- Praise, 2026-11-01, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “just automated my bookkeeping with claude duet. it monitors my bank, email, virtual cards, writes the ledger, puts invoices / receipts in a folder and links to them from the ledger. runs once per month.” [source](https://twitter.com/1688359509826105345/status/2117029674551513525)
- Complaint, 2026-11-08, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “okay real talk, i tried claude duet and i am not sold yet 👀 it feels a little clunky to me, like it tries to do everything but i still end up fixing stuff myself maybe i am using it wrong though, so tell me... are you using it?? what do you actually use it for??” [source](https://twitter.com/1760272696967856128/status/2119417119636926708)
- Complaint, 2026-10-31, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “thanks claude duet for not sending the important email and just sitting on it for days [link] [link]” [source](https://twitter.com/1294737900236681223/status/2116584913319420290)
- Complaint, 2026-10-26, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “i've had claude duet for about a week. i've barely opened it. i asked if i could direct all my claude projects from it. it can read them. it can't run work inside them. has anyone gotten grok bot or openai's dots to actually run their projects? [link]” [source](https://twitter.com/335276818/status/2114764309779906586)

### Running on its own: 6th of 10 (customer love 0.481, n 49)

Working on its own comes up 1.4 times as often as for other agents. Praise outnumbers complaints, and love stays within the category norm. In posts about price, users say these background runs eat into their usage.

| Criterion | Reading | Customer love | 95% interval | n | Praise | Complaint |
|---|---|---|---|---|---|---|
| [Unprompted suggestions, insights and check-ins](https://feedbackbench.com/criteria/auto.proactive.md) | Too few posts | 0.509 | 0.485–0.530 | 26 | 23 | 3 |
| [Remembering the user and context across sessions](https://feedbackbench.com/criteria/auto.memory.md) | Too few posts | 0.498 | 0.484–0.511 | 10 | 6 | 4 |
| [Multiple agents, teams and delegation between them](https://feedbackbench.com/criteria/auto.multi_agent.md) | Too few posts | 0.481 | 0.457–0.502 | 8 | 4 | 4 |
| [Where the agent runs: its own computer and always-on running](https://feedbackbench.com/criteria/auto.cloud_env.md) | Too few posts | 0.487 | 0.471–0.501 | 8 | 3 | 5 |
| [Scheduled recurring tasks and daily briefings run](https://feedbackbench.com/criteria/auto.routines.md) | Too few posts | 0.506 | 0.501–0.512 | 5 | 5 | 0 |
| [Reminders and notifications reach the user correctly](https://feedbackbench.com/criteria/auto.reminders.md) | Too few posts | 0.491 | 0.471–0.504 | 4 | 2 | 2 |

Most recent posts:

- Praise, 2026-11-08, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “the most noteworthy differentiator of claude duet is how its excellent taste for when to be proactive. the others can be a little tedious.” [source](https://twitter.com/1121438638125080579/status/2119462612064752105)
- Praise, 2026-11-04, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “ok claude duet was very proactive and intelligent here. sent within the hour i got the email. [link]” [source](https://twitter.com/812741641081536512/status/2117801468698697769)
- Praise, 2026-11-04, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “moving many of my grok @bot routines over to claude duet. so far so good.” [source](https://twitter.com/1462761/status/2117803019496804559)
- Complaint, 2026-10-28, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “i tried claude duet. after weeks with grok bot and meta muse, it feels like an afterthought. every assistant tries to be proactive. none truly is yet. the real work: budget, cost, and beating what loops, flywheels and agents already deliver. duet doesn't move that needle. [link]” [source](https://twitter.com/96748069/status/2115432540622532647)
- Complaint, 2026-11-06, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “either anthropic has not fully implemented the "claude duet messaging code sessions" feature yet, or they're just trolling at this point. in what context would i want to have duet message a session (for which i have to tap an allow card) when it can't even get notified if the session has replied? the only way for duet to know the session has responded/finished is for me to tell it, which requires…” [source](https://twitter.com/1864552063524306944/status/2118758529855054310)
- Complaint, 2026-10-30, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “my thoughts on claude duet vs gpt dots. dots aren't perfect, but they're pretty helpful. they actually are proactive, create sessions, don't intrinsically count against my subscription use, and i have found helpful in my day to day life. claude duet can't actually do anything useful for me. everything i've asked it to do, it says it can't. core to this is the fact that it can't actually launch new…” [source](https://twitter.com/1634388854965665793/status/2116168780426391968)

### Staying in control: 5th of 8 (customer love 0.513, n 30)

Users talk about control 1.79 times as often as for other agents. Love here is 0.513, the highest of any area, and within the category norm.

| Criterion | Reading | Customer love | 95% interval | n | Praise | Complaint |
|---|---|---|---|---|---|---|
| [Which AI model runs, how good it is, and choosing it](https://feedbackbench.com/criteria/control.model.md) | Too few posts | 0.522 | 0.510–0.536 | 17 | 15 | 2 |
| [Asks before acting: too little or too much](https://feedbackbench.com/criteria/control.approval.md) | Too few posts | 0.506 | 0.494–0.518 | 9 | 6 | 3 |
| [Seeing what the agent is doing and has done](https://feedbackbench.com/criteria/control.visibility.md) | Too few posts | 0.493 | 0.483–0.502 | 6 | 1 | 5 |
| [Configuring behaviour, skills and portability](https://feedbackbench.com/criteria/control.customize.md) | Too few posts | 0.496 | 0.481–0.508 | 5 | 3 | 2 |

Most recent posts:

- Praise, 2026-11-02, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “could my new ai agent email a client without asking me? gave claude duet a test invoice and one instruction: draft a follow-up. never said send. it wrote the email and stopped on a not now / send card. nothing went out until i tapped send. the approval gate held. [link]” [source](https://twitter.com/2057903995968954368/status/2117346272625487989)
- Praise, 2026-11-06, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “@walrus_systems this claude duet memory graph is insane. real chief of staff energy, but it actually asks first” [source](https://twitter.com/1501686387774476295/status/2118798273460732385)
- Praise, 2026-10-31, Grok Bot + Claude Code, Why is Everyone Switching to This? (YouTube): “claude duet > t3 > open source grok bot (codync) on my own server. i’m just pushing approve and making sanity checks. the babysitting era is finally over” [source](https://www.youtube.com/watch?v=to4Me479A2s&lc=UgxQskxaSThKqKhBuOx4AaABAg)
- Complaint, 2026-10-29, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “@gabemulley @dayahdover dm'd you again gabe, sorry.. but claude duet can't do a simple fix and deploy without asking for authorization like 5 times and it still hasn't finished. it's stuck in this authorization loop for some reason. it was a simple fix i asked for” [source](https://twitter.com/128274666/status/2115833599719288950)
- Complaint, 2026-10-25, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “claude duet got stopped by medium's bot check, and i had no way into its browser to help. that was the end of it same task, 3 agents "sign into medium with an email code from my gmail, ask me before using the code, tell me my profile name" grok bot: done, asked before using the code muse: done. made me save my login first, hit a bot check, got through on its own, asked before using the code claude…” [source](https://twitter.com/1628114379958026240/status/2114465161629036903)
- Complaint, 2026-11-03, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “finally got into claude duet. i love claude code, but duet isn't clicking. it’s too siloed from the main app, too many threads and sub threads, and i never know what it's actually doing. not much more agentic than cowork. how are you finding it?” [source](https://twitter.com/14977595/status/2117474068673020400)

### Trusting it with data and safety: Too few posts (customer love 0.491, n 6)

Users raise privacy and safety less often than for other agents, 0.62 times the rate. There are too few posts to rank. Every user-week here is a complaint.

| Criterion | Reading | Customer love | 95% interval | n | Praise | Complaint |
|---|---|---|---|---|---|---|
| [Refuses legitimate requests because of guardrails](https://feedbackbench.com/criteria/trust.refusals.md) | Too few posts | 0.494 | 0.488–0.499 | 4 | 0 | 4 |
| [Permission scope requested and isolation between agents](https://feedbackbench.com/criteria/trust.access_scope.md) | Too few posts | 0.498 | 0.494–0.500 | 1 | 0 | 1 |
| [Vulnerabilities, prompt injection and cross-user leaks](https://feedbackbench.com/criteria/trust.security.md) | Too few posts | 0.498 | 0.495–0.500 | 1 | 0 | 1 |
| [How the vendor collects, keeps and uses data](https://feedbackbench.com/criteria/trust.data_use.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |

Most recent posts:

- Complaint, 2026-10-28, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “claude duet sure does want a lot of - if not complete - access to my desktop.” [source](https://twitter.com/13160652/status/2115504937362665946)

### Dealing with the vendor: Too few posts (customer love 0.507, n 3)

Users rarely mention updates and support, at 0.13 times the rate for other agents. There are too few posts to rank.

| Criterion | Reading | Customer love | 95% interval | n | Praise | Complaint |
|---|---|---|---|---|---|---|
| [Got better or worse over time, and update cadence](https://feedbackbench.com/criteria/vendor.updates.md) | Too few posts | 0.506 | 0.500–0.516 | 2 | 2 | 0 |
| [Support responses and account bans or lockouts](https://feedbackbench.com/criteria/vendor.support.md) | Too few posts | 0.499 | 0.496–0.500 | 1 | 0 | 1 |
| [Outages, errors and agent going unresponsive](https://feedbackbench.com/criteria/vendor.uptime.md) | Too few posts | 0.500 | 0.500–0.500 | 0 | 0 | 0 |

Most recent posts:

- Praise, 2026-10-26, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “claude duet new app update 👀 just read the notes and wow the feedback i sent today was just added! 🥳 [link]” [source](https://twitter.com/1786846062390431744/status/2114554472668049717)
- Praise, 2026-10-25, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “@iruletheworldmo and claude duet is already making massive improvements every new testflight update and that fucker is still in early beta. i just hope they keep this trend up. openai was doing decent for a while then they got full of themselves and now… well, now they have dots and it’s 💩” [source](https://twitter.com/1422719919418920966/status/2114460551694954573)
- Complaint, 2026-11-02, X search: Claude Duet, ClaudeDuet, Anthropic Duet, Anthropic's Duet, Duet by Anthropic (X): “don't apply for that claude duet thing. i carelessly filled in the information to join the waitlist today, and my account got banned in less than two hours. losing over 200 dollars on the subscription really hurts. the key is that several projects were halfway done. sigh 😮💨” [source](https://twitter.com/371777666/status/2117267826788675736)
