Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Claude Tag

anthropic.com

Post reply on HN
Print view View original

191–200 of 200 posts

Re: Claude Tag

#191

Post by port11 »

skywhopper wrote: post #185
munksbeer wrote: post #181

Earlier quoted context omitted.

I keep seeing this sentiment. Claude Code cli works very well for me. To lay my cards out, I think the negativity towards it isn't driven by anything objective, but by anti-AI / anti-big-tech sentiment.

I’d say you are wrong about the “anti-AI sentiment”. It’s possible to recognize the tool works well in general and also objectively observe as it adds contradictory features and evolves behavior in ways to encourage token waste. Claude Code CLI is also a very narrow part of their portfolio at the moment.

CC can waste reams of tokens indeed.

Pi with hash-based replaces is very token-efficient. That plus cache headers for OpenCode Go and here I am, unable to hit the 5-hour limit unless I use GLM (which doesn’t cache).

Anthropic’s incentive to save tokens on subscription users directly clashed with their incentive to waste tokens on Enterprise.

Re: Claude Tag

#193

Post by linhns »

dbbk wrote: post #188
munksbeer wrote: post #181

Earlier quoted context omitted.

I keep seeing this sentiment. Claude Code cli works very well for me. To lay my cards out, I think the negativity towards it isn't driven by anything objective, but by anti-AI / anti-big-tech sentiment.

No, their products are just janky and unreliable. Have you tried using their cloud environments for Claude Code? They're unusable, and I'm not choosing that word lightly. There are even blatant UX issues in the consumer Claude app that have gone unfixed for months and months and months. For example, if Claude asks you a question and it runs over 2 lines? It's truncated and you can't read it. This would never pass QA,…

I agree on the UX part, sometimes I have scrolling or blank screen when exiting. Cloud envs have never been CC's target tbf, and I'm happy how it works in the terminal.

Re: Claude Tag

#195

Post by disillusioned »

kylecazar wrote: post #132
SAK_ATAK wrote: post #98

I don't understand how this is gonna fly for enterprise security and compliance. Claude needs to inherit permissions from somewhere, and those permissions will never align with the members of a slack channel. And finding the lowest common denominator of access probably results in a dumbed-down, useless experience. The only way it works is if customers truly start treating agents as humans with the same liability as a…

An admin scopes permissions on a per-channel basis. It doesn't allow external actions until an owner specifically provisions that tool for that channel. I think. But people can be invited to a channel after @Claude is provisioned. So yeah, I suppose you'll need to be deliberate about channel memberships.

This sounds lovely, and ideal, but my read on this situation for, say, Google Drive, is a bit more nuanced: Anthropic's docs say that its permissions are scoped to the folders you grant access to the _user_ you create for Claude.

This creates an unfortunate situation where, for example, I can't scope a given channel's Claude Tag to _only_, say, a project's shared drive or specific folders, because I'm using a custom claude@ Google account in my org, and its permissions have to be all-encompassing.

It also means I can't escalate privilege and expose certain documents that I would like to: say I have a private #hr channel that I would like to have access to our HR shared drive docs with proprietary information. My understanding is that any user in any channel with @claude tag in it would be able to interrogate Claude about any file that the Claude Google user has access to, regardless of which channel they're in.

I'm trying to determine if the way around this is the alternate Google service account option, but that mentions domain-wide delegation in such a manner that it makes me think it would replicate the problem, and I obviously don't want to create custom Google accounts for each project scope in my org...

Re: Claude Tag

#196

Post by jessinra98 »

Tag's memory is Slack specific which makes sense, that's the product. But I spend as much time in Claude Code and Cursor as I do in Slack. Been running an open-source MCP memory server (lorekeeper, https://github.com/jessinra/lorekeeper) and the surprise is one install just works across all of them, same memory graph. Makes me think decoupling memory from the agent is the way to go. Anyone else doing cross-tool memory?

Re: Claude Tag

#197

Post by austinxu »

insane_dreamer wrote: post #96

Not on Teams, I see. Likely by design :) Not sure how much we'll use this, but it could be useful for filing tickets from conversations. Though I'd prefer to just point claude to the convo post-hoc rather than have to invite it each time just in case I want to ask claude to do something. The last thing I want is claude chipping into a convo Clippy-style.

[dead]

Re: Claude Tag

#198

Post by edihasaj »

jessinra98 wrote: post #196

Tag's memory is Slack specific which makes sense, that's the product. But I spend as much time in Claude Code and Cursor as I do in Slack. Been running an open-source MCP memory server (lorekeeper, https://github.com/jessinra/lorekeeper ) and the surprise is one install just works across all of them, same memory graph. Makes me think decoupling memory from the agent is the way to go. Anyone else doing cross-tool memo…

[flagged]

Re: Claude Tag

#199

Post by mogili »

mhitza wrote: post #170
mogili wrote: post #164

Earlier quoted context omitted.

Token consumption on this is not a huge issue, I built this a few months ago for our company slack using claude agent sdk and most of the conversations/sessions were short lived. You don't generally tend to have long sessions with agents in Slack like you would if using in the terminal. It was very useful to delegate things like create a ticket based on a discussion to Claude.

> You don't generally tend to have long sessions with agents in Slack like you would if using in the terminal. You have to check the announcement fineprint. As long as they advertise it as proactive it needs to keep up to date with all the ongoing discussions. And I believe that means all group messages. Wouldn't that make sense? From the announcement > follows along with its channel, it builds more context about the…

It means it'll use tool calls to search slack and get context not that it's some sort of ambient always running loop.

Re: Claude Tag

#200

Post by mogili »

mogili wrote: post #199
mhitza wrote: post #170

Earlier quoted context omitted.

> You don't generally tend to have long sessions with agents in Slack like you would if using in the terminal. You have to check the announcement fineprint. As long as they advertise it as proactive it needs to keep up to date with all the ongoing discussions. And I believe that means all group messages. Wouldn't that make sense? From the announcement > follows along with its channel, it builds more context about the…

It means it'll use tool calls to search slack and get context not that it's some sort of ambient always running loop.

They do have an ambient mode you can enable.
Post reply on HN