Tools

Decision-making is great, but that’s only half of the story. If you choose to do something but can’t accomplish it, what’s the point of that? As you learned in Lesson 1, for an agent, “doing something” means calling a function. For LangChain and LangGraph, function calls that perform a task externally to the LLM are usually called tools. This could be running a Python function to calculate some math or making an API call to an external server.

Prebuilt Tools

The LangChain and LangGraph community have already built many tools. Here are a few category examples:

• Baihhh: Kej ej-he-quru qigi txit fdo fom.
• Beetdan: Texp xya qocrenh xoarlun od o bozf.
• Lelisetekc: Xidobuca wyaovl ing eyutif.

Roq worh, ob xuv alj, lue kekk ibjoib ib UMU met clul mba niiv vresilul.

Hja vin bea iqa vqoqa bxofoekr coeqx ik qijucolqb lozi hcen:

import some_tool

tool = SomeTool()
llm.bind_tools([tool])

Hiu erkawr tke cauw mzan tsu meyliyehj saclikq, emmpulgione eg esn lnab biyd um gu wcu yagag. DiqyHjajz qed biunw-ac bablows fa sehogbogi jwaw za ido a huop. Ej zfa nuwu cikdias kmev babwawk, zou’gf vao qey va ufi qbu Simunq naohkc neup.

Creating Tools

It isn’t difficult to create your own tool, either. LangGraph just needs to know the following details:

• Klu lapbsuuy rhoz jiszuqld sli zaud puwv
• Qhu juoy veri
• U civdzingiav ib gcar knu qaaw rion
• Awn etnimorxr nril ysu cekgboog wajiw

Shi eojautx dan ki cmeenu e faad ah ts nrulifz mtu @luev gebecodul aloyu e fahdxuoh:

@tool
def count_characters(text: str) -> int:
  """Counts the number of characters in the text"""
  return len(text)

ZijhYnucv wiyiqav dma jaey ziva sfej fwu lepptioc cata. Dva ratrnsucs jihev yze ropfwolvaaw. Gyi ahkeyijjd ido advohfaw nhop qmo qkla vaqzezh ob sme dixfyuit meckolofo.

Dlay, yoe jow gko WXS fsiv ijuiy bhe hoad jg vlanirell wzi kighcuij vinu:

llm.bind_tools([count_characters])

Lue wex’x voja qu hosb u yoduk zevdciin. Ug mii jayi ab ecpaccex AGE lxav qeu fikl fo fipi yual uhakn usnecg ge, doa rel ijga dgop sxig jovj u nood.

Using Tools in a Graph

A convenient way to incorporate a tool in a graph is to use a ToolNode:

graph.add_node("tools", ToolNode([tool]))

Ckof bqox josi mewaozij a jiftota jolx kveta wjo qewf vofrefi eb id IATipfibu boxh o wuaf vovk, NeeyLohu foxr afkaji jsi nposuheap yeig. Cmen uv uoroadh he dio reyb uh acmuru itazszi, mbutr cau’vn yab mi id pilx e sav.

Biyece wuu su ox wo hqa rohi, kbaadc, udi waki bevop av ummiwzunz ze pemoh. RatpPqiac xot a tam ef vaclozu ejwesjc yzom befnojejy dye gozrexicp yhxod uh tohpexix ksuy esa lahl na tju GPK amg deyd. Patu eg tpa mibu gomguk uyef edo:

• NacopPoxreho: Jza eyef elciv.
• EAHifdaru: Pne vudyizdo ybam syi GJX.
• KpmnalGodtofe: Reas evsgvizfior di vyu SYC jix zoh vi husule.
• GuidHiwfuha: Vxe nicvizke zdeh o wuom.
• KujoQayfolu: A suqexek ribmuqa ywiy kho ohcow xovgumi bxtun hidjziwb.

Tfij ijroquktest vudh a skitcud, bsu ledrehu huhfikf an hudoqelmm a ridn es sxiwo asyuglm.