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

261–270 of 280 posts

Re: Claude Advanced Tool Use

#261

Post by zbowling »

cube2222 wrote: post #5

Nice! Feature #2 here is basically an implementation of the “write code to call tools instead of calling them directly” that was a big topic of conversation recently. It uses their Python sandbox, is available via API, and exposes the tool calls themselves as normal tool calls to the API client - should be really simple to use! Batch tool calling has been a game-changer for the AI assistant we've built into our produ…

I wrote a better version of this idea: https://github.com/zbowling/mcpcodeserver

It works as an MCP proxy of sorts that converts all the child MCP tools into typescript annotations, asks your LLM to generate typescript, then executes those tool calls in a restricted VM to do the tool calls that way. It allows parellel process, passing data between tools without coming back to the LLM for a full loop, etc. The agents are pretty good at debugging issues they create too and trying again.

Re: Claude Advanced Tool Use

#262

Post by zbowling »

rfw300 wrote: post #15

I am extremely excited to use programmatic tool use. This has, to date, been the most frustrating aspect of MCP-style tools for me: if some analysis requires the LLM to first fetch data and then write code to analyze it, the LLM is forced to manually copy a representation of the data into its interpreter. Programmatic tool use feels like the way it always should have worked, and where agents seem to be going more bro…

I built a MCP server that solves this actually. It works like a tool calling proxy that calls child servers but instead of serving them up as direct tool calls, it exposes them as typescript defintions, asks your LLM to write code to invoke them all together, and then executes that typescript in a restricted VM to do tool calling indirectly. If you have tools that pass data between each other or need some kind of parsing or manipulation of output, like the tool call returns json, it's trivial to transform it. https://github.com/zbowling/mcpcodeserver

Re: Claude Advanced Tool Use

#263

Post by slimslenders »

michaelanckaert wrote: post #23

The "Tool Search Tool" is like a clever addition that could easily be added yourself to other models / providers. I did something similar with a couple of agents I wrote. First LLM Call: only pass the "search tool" tool. The output of that tool is a list of suitable tools the LLM searched for. Second LLM Call: pass the additional tools that were returned by the "search tool" tool.

I think this is very true. Tool search tools can be model agnostic. And programmatic tool calling really just needs a code sandbox tool. We've provided some examples of these patterns on top of a local docker engine (oss project is here https://github.com/docker/mcp-gateway/ and blog is https://www.docker.com/blog/dynamic-mcps-stop-hardcoding-you...).

Re: Claude Advanced Tool Use

#264

Post by ewoodrich »

nautilus12 wrote: post #220
jimbo808 wrote: post #154

Earlier quoted context omitted.

Claude is pretty good at totally disregarding most of what’s in your CLAUDE.md, so I’m not optimistic. For example a project I work on gives it specific scripts to run when it runs automated tests, because the project is set up in a way that requires some special things to happen before tests will work correctly. I’ve never once seen it actually call those scripts on the first try. It always tries to run them using t…

I had the same problem. My Claude md eventually gets forgotten and it forgets best practices that I put in there. I've switched to using hooks that run it through a variety of things like requiring testing. That seems to work better than Claude md because it has to run the hook every time it makes changes.

I really need something like this up for tasks I want Claude to run before handing off a task to me as "complete". It routinely ignores my instructions of checklist items that need to be satisfied to be considered successful. I have a helper script documented in CLAUDE.md that lets Claude or me get specific build/log outputs with a few one liner commands yet Claude can't be bothered to remember running them half the time.

Way too frequently Claude goes, "The task is fully implemented, error free with tests passing and no bugs or issues!" and I have to reply "did you verify server build/log outputs with run-dev per CLAUDE.md". It immediately knows the command I am referencing from the instructions buried in its context already, notices an issue and then goes back and fixes it correctly the second time. Whenever it happens it instantly makes an agentic coding session go from feeling like breezy, effortless fun to pulling teeth.

I've started to design a subagent to handle chores after every task to avoid context pollution but it sounds like hooks are the missing piece I need to deterministically guarantee it will run every time instead of just when Claude feels the vibes are right.

Re: Claude Advanced Tool Use

#267

Post by airspresso »

jimbo808 wrote: post #154
losvedir wrote: post #80

I never really understood why you have to stuff all the tools in the context. Is there something wrong with having all your tools in, say, a markdown file, and having a subagent read it with a description of the problem at hand and returning just the tool needed at that moment? Is that what this tool search is?

Claude is pretty good at totally disregarding most of what’s in your CLAUDE.md, so I’m not optimistic. For example a project I work on gives it specific scripts to run when it runs automated tests, because the project is set up in a way that requires some special things to happen before tests will work correctly. I’ve never once seen it actually call those scripts on the first try. It always tries to run them using t…

Sounds like you're fighting the weights. What would it take to align the setup with what the LLM expects?

Re: Claude Advanced Tool Use

#268

Post by PanMan »

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…

Jira actually has both an MCP server and a CLI tool (called "acli"). I switched our claude code to the CLI (with a skill) from the MCP as it seems.. more efficient/quicker.

Re: Claude Advanced Tool Use

#269

Post by ElasticBottle »

zeroq wrote: post #162
checker659 wrote: post #153

Earlier quoted context omitted.

This is the most creative comment I've read on HN as of late.

Thanks, most of the times when I do that people tell me to stop being silly and stop saying nonsense. ¯\_(ツ)_/¯

don't listen to the naysayers! Had a chuckle as well LoL

Re: Claude Advanced Tool Use

#270

Post by cube2222 »

zbowling wrote: post #261
cube2222 wrote: post #5

Nice! Feature #2 here is basically an implementation of the “write code to call tools instead of calling them directly” that was a big topic of conversation recently. It uses their Python sandbox, is available via API, and exposes the tool calls themselves as normal tool calls to the API client - should be really simple to use! Batch tool calling has been a game-changer for the AI assistant we've built into our produ…

I wrote a better version of this idea: https://github.com/zbowling/mcpcodeserver It works as an MCP proxy of sorts that converts all the child MCP tools into typescript annotations, asks your LLM to generate typescript, then executes those tool calls in a restricted VM to do the tool calls that way. It allows parellel process, passing data between tools without coming back to the LLM for a full loop, etc. The agents…

Could you expand in what way it’s better?

So far what you described sounds like what they did, but they manage the sandboxed environment for me and use Python rather than TypeScript.

Do note that their thing works not only with MCP tools, but arbitrary tools.

Post reply on HN