Add core implementation for AI-powered question answering

Introduce multiple agents, tools, and utilities for processing, extracting, and answering user-provided questions using LLMs and external data. Key features include knowledge processing, question splitting, search term generation, and contextual knowledge handling.

pkg/agents/tools/wolfram.go

@@ -0,0 +1,34 @@
+package tools
+
+import (
+	"github.com/Edw590/go-wolfram"
+
+	gollm "gitea.stevedudenhoeffer.com/steve/go-llm"
+)
+
+type WolframFunctions struct {
+	Imperial *gollm.Function
+	Metric   *gollm.Function
+}
+
+func CreateWolframFunctions(appId string) WolframFunctions {
+	client := &wolfram.Client{AppID: appId}
+	return WolframFunctions{
+		Imperial: gollm.NewFunction(
+			"wolfram",
+			"Query the Wolfram Alpha API",
+			func(ctx *gollm.Context, args struct {
+				Question string `description:"The question to ask Wolfram|Alpha"`
+			}) (string, error) {
+				return client.GetShortAnswerQuery(args.Question, wolfram.Imperial, 10)
+			}),
+		Metric: gollm.NewFunction(
+			"wolfram",
+			"Query the Wolfram Alpha API",
+			func(ctx *gollm.Context, args struct {
+				Question string `description:"The question to ask Wolfram|Alpha"`
+			}) (string, error) {
+				return client.GetShortAnswerQuery(args.Question, wolfram.Metric, 10)
+			}),
+	}
+}