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

Claude Advanced Tool Use

anthropic.com

Post reply on HN
Print view View original

201–210 of 280 posts

Re: Claude Advanced Tool Use

#201

Post by emilsoman »

vanviegen wrote: post #180
emilsoman wrote: post #177

> The script runs in the Code Execution tool (a sandboxed environment), pausing when it needs results from your tools. When you return tool results via the API, they're processed by the script rather than consumed by the model. The script continues executing, and Claude only sees the final output. Anyone knows how they would have implemented the pause/resume functionality in the code execution sandbox? I can think of…

Presumably, a tool call is just a library call in the script. The implementation would need to ask the environment outside the sandbox (through a socket?) to take some action on its behalf.

That's cool, but they say the code execution would wait till the tool call is done. Would they be just keeping the code execution process alive? That seems like a bad idea given tool calls can take an unknown amount of time to finish. I am guessing they would be putting the python orchestrator code to sleep when the tool call starts and restoring the state when the tool call is done.

Re: Claude Advanced Tool Use

#202

Post by theknarf »

We should just build more CLI tools, that way the agentic AI can just run `yourtool --help` to learn how to use it. Instead of needing an MCP-server to access ex. Jira it should just call a cli tool `jira`. Better CLI tools for everything would help both AI and humans alike.

Re: Claude Advanced Tool Use

#203

Post by _pdp_ »

fy20 wrote: post #168
_pdp_ wrote: post #24

Our agentic builder has a single tool. It is called graphql. The agent writes a query and executes it. If the agent does not know how to do particular type of query then it can use graphql introspection. The agent only receives the minimal amount of data as per the graphql query saving valuable tokens. It works better! Not only we don't need to load 50+ tools (our entire SDK) but it also solves the N+1 problem when u…

Whoa there, you don't need to be so sadistic to your team. It's not GraphQL, but having a document describing how your API works, including types, that is important. I expect you could achieve the same with a comprehensive OpenAPI specification. If you want something a bit stricter I guess SOAP would work too, LLMs love XML after all.

We have well described OpenAPI and GraphQL specifications already. :)

Being AI-first means we are naturally aligned with that kind of structured documentation. It helps both humans and robots.

Re: Claude Advanced Tool Use

#204

Post by _pdp_ »

bravura wrote: post #92
_pdp_ wrote: post #24

Our agentic builder has a single tool. It is called graphql. The agent writes a query and executes it. If the agent does not know how to do particular type of query then it can use graphql introspection. The agent only receives the minimal amount of data as per the graphql query saving valuable tokens. It works better! Not only we don't need to load 50+ tools (our entire SDK) but it also solves the N+1 problem when u…

Isn't the challenge that introspecting graphql will lead to either a) a very long set of definitions consuming many tokens or b) many calls to drill into the introspection?

Well either that or stuff the tool usage examples into the prompt for every single request. If you have only 2-3 tools GraphQL is certainly not necessary - but it wont blow up the context either. If you have 50+ tools, I don't see any other way to be honest, unless you create your own tool discovery solution - which is what GraphQL does really well with the caveat that whatever you decide to do is certainly not natural to these LLMs.

Keep in mind that all LLMs are trained on many GraphQL examples because the technology has been in existence since 2015. While anything custom might just work it is certainly not part of the model training set unless you fine-tune.

So yes, if I need to decide on formats I will go for GraphQL, SQL and Markdown.

Re: Claude Advanced Tool Use

#206

Post by tomashubelbauer »

theknarf wrote: post #202

We should just build more CLI tools, that way the agentic AI can just run `yourtool --help` to learn how to use it. Instead of needing an MCP-server to access ex. Jira it should just call a cli tool `jira`. Better CLI tools for everything would help both AI and humans alike.

This would be awesome, but great CLIs would have already been valuable prior to the age of LLMs and yet most services didn't ship one. I think it is because services like Jira and others do not want to be too open. Ultimately, despite the current LLM/MCP craze, I think this won't change and MCP tools will start getting locked down and nerfed somehow, the same way APIs have in not so recent memory after there being a bit of a craze around those a decade+ back.

Re: Claude Advanced Tool Use

#207

Post by ath92 »

tomashubelbauer wrote: post #206
theknarf wrote: post #202

We should just build more CLI tools, that way the agentic AI can just run `yourtool --help` to learn how to use it. Instead of needing an MCP-server to access ex. Jira it should just call a cli tool `jira`. Better CLI tools for everything would help both AI and humans alike.

This would be awesome, but great CLIs would have already been valuable prior to the age of LLMs and yet most services didn't ship one. I think it is because services like Jira and others do not want to be too open. Ultimately, despite the current LLM/MCP craze, I think this won't change and MCP tools will start getting locked down and nerfed somehow, the same way APIs have in not so recent memory after there being a…

Nobody shipped this because previously almost nobody could use CLI tools. Now you can just ask an llm to generate the commands which makes things much more accessible

Re: Claude Advanced Tool Use

#208

Post by delaminator »

theknarf wrote: post #202

We should just build more CLI tools, that way the agentic AI can just run `yourtool --help` to learn how to use it. Instead of needing an MCP-server to access ex. Jira it should just call a cli tool `jira`. Better CLI tools for everything would help both AI and humans alike.

That's only useful if the agent is running in your terminal. The example given about updating a cell in Excel, I mean, I suppose that is a sort of tool you could used for something. SharePoint has an API for updating excels on SharePoint. But to update a single cell is actually quite time consuming for the API round trip - multiple seconds. I recently had to rewrite something because it was doing individual API calls to update cells.

Re: Claude Advanced Tool Use

#209

Post by delaminator »

fny wrote: post #74
jmward01 wrote: post #28

The Programmatic Tool Calling has been an obvious next step for a while. It is clear we are heading towards code as a language for LLMs so defining that language is very important. But I'm not convinced of tool search. Good context engineering leaves the tools you will need so adding a search if you are going to use all of them is just more overhead. What is needed is a more compact tool definition language like, I d…

Why exactly do we need a new language? The agents I write get access to a subset of the Python SDK (i.e. non-destructive), packages, and custom functions. All this ceremony around tools and pseudo-RPC seems pointless given LLMs are extremely capable of assembling code by themselves.

I'm imagining something more like Rexx with quite high level commands. But that certainly blurs the line between programming language and shell.

The reason for choosing higher level constructs is token use. We certainly reduce the number of tokens by using a shell like command language, But of course that also reduces expressiveness.

I've been meaning to get round to Plan 9 style where the LLM reads and writes from files rather than running commands. I'm not sure whether that's going to be more useful than just running commands. is for an end user because they only have to think about one paradigm - reading/writing files.

Re: Claude Advanced Tool Use

#210

Post by delaminator »

sandbags wrote: post #193
ctoth wrote: post #135

Earlier quoted context omitted.

I'm not sure that we need a new language so much as just primitives from AI gamedev, like behavior trees along with the core agentic loop.

After implementing a behaviour tree library and realising the power of select & sequence I found myself wondering why they aren’t used more widely. I’ve never done anything in crypto but watched in horror as people created immutable contracts with essentially Javascript programs. Surely it would be much easier to reason about/verify scripts written as a behaviour tree with a library of queries and actions. Even being…

Seeing as it was your inspiration, here is a summary of a discussion with Claude on this topic. (not the crypto part.)

https://claude.ai/public/artifacts/2b23f156-c9b5-42df-9a83-f...

Post reply on HN