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

251–260 of 280 posts

Re: Claude Advanced Tool Use

#251

Post by misiti3780 »

chillfox wrote: post #248
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 pretty much how I have been using coding agents. I get them to build small cli tools with a --help option and place them in a `./tools` directory. Then I can tell an agent to use the tools to accomplish whatever task I need done. Usually when I mention a `tool --help` for the first time in a prompt I will put it in backticks with the --help argument. It works really well.

im going to try this. it sounds promising. can you provide an example for more context?

Re: Claude Advanced Tool Use

#252

Post by lewisjoe »

The criticisms here surprise me. "Programmatic Tool Calling" is a huge leap when you want AI to work with your app - like a human would.

I've been trying to get LLMs to work in our word processor documents like a human collaborator following instructions. Writing a coding agent is far more straightforward (all code are just plain strings) than getting an agent to work with rich text documents.

I imagined the only sane way is to expose a document SDK and expect AI to write programs that call those SDK APIs. That was the only way to avoid MCPs and context explosion. Claude has now made this possible and it's exciting!

Hope the other AI folks adopt this as well.

Re: Claude Advanced Tool Use

#253

Post by exographicskip »

prescriptivist wrote: post #149
JoshGlazebrook wrote: post #99

Is there a good guide for all of these concepts in claude code for someone coming from Cursor? I just feel like the amount of configuration is overwhelming vs. Cursor to accomplish the same things.

Most guides to wringing productivity out of these higher level Claude code abstractions suffer from conceptual and wall-of-text overload. Maybe it's unavoidable but it's tough to really dig into these things. One of the things that bugs me about AI-first software development is it seems to have swung the pendulum of "software engineering is riddled with terrible documentation" to "software engineering is riddled with…

Thanks for the new word re: prolix! Couldn't quite pin down why heavily AI generated posts/documentation felt off — aside from an amorphous _feeling_ — until today.

Re: Claude Advanced Tool Use

#254

Post by zby »

There is huge difference between tools executed on the client and those that run on the server - I wish it was made more clear in announcements like this one what it is referring to.

Re: Claude Advanced Tool Use

#255

Post by PantaloonFlames »

easyascake wrote: post #238
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.

I use the GitLab CLI (glab) extensively, because it is so much better than the (official) GitLab MCP. I just run `glab auth login` before launching Claude Code, then tell CC to use `glab` to communicate with the GitLab API. When using the MCP, I have to do a whole OAuth browser-launch process and even then I am only limited to the 9-10 tools that they've shipped it with so far.

See also “95% of MCP Servers are useless”. https://youtu.be/7baGJ1bC9zE?si=ShyLg2mHWwbBW1DS

tl;dr AI-powered assistants can already use command line tools.

Re: Claude Advanced Tool Use

#256

Post by PantaloonFlames »

mlrtime wrote: post #214
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 is exactly what I do when given an PAT + api documentation, write my own tool. Sure it'd be better if atlassian did it, but I'm not holding my breath.

Or, why not let the LLM write the tool and give it to the agent? Taking it one step further the tool could be completely ephemeral - it could have a lifetime of exactly one chat conversation.

Re: Claude Advanced Tool Use

#257

Post by doctorpangloss »

ACCount37 wrote: post #198
jawns wrote: post #77

I'm starting to notice a pattern with these AI assistants. Scenario: I realize that the recommended way to do something with the available tools is inefficient, so I implement it myself in a much more efficient way. Then, 2-3 months later, new tools come out to make all my work moot. I guess it's the price of living on the cutting edge.

The consequences of having the world's smartest people working on those things 24/7. Often, either the model itself gets improvements that render past scaffolding redundant, or your clever hacks to squeeze more performance out get obsoleted by official features that do the same thing better.

> The consequences of having the world's smartest people working on those things 24/7.

haha, don't you worry, they are going to be back to working on ads - inside the chatbots - soon enough

Re: Claude Advanced Tool Use

#258

Post by 3acctforcom »

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.

JIRA is a great example. I used to have automation when it was hosted on prem and I had database access.

Now it's locked into the cloud with piss poor APIs so they can sell you more add-ons. I'm actively looking at alternatives.

Re: Claude Advanced Tool Use

#259

Post by tacone »

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.

And cli tools can be composed with scripts, which makes the whole experience faster and reproducible.

Re: Claude Advanced Tool Use

#260

Post by zbowling »

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…

I specifically built this as an MCP server. It works like an MCP server that proxies to other MCP servers and converts the tool defintions in to typescript anotations and asks your llm to generate typescript that runs in a restricted VM to make tools calls that way. It's based on the apple white paper on this topic from last year. https://github.com/zbowling/mcpcodeserver
Post reply on HN