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/calculator.go

@@ -0,0 +1,27 @@
+package tools
+
+import (
+	"go.starlark.net/lib/math"
+	"go.starlark.net/starlark"
+	"go.starlark.net/syntax"
+
+	gollm "gitea.stevedudenhoeffer.com/steve/go-llm"
+)
+
+var Calculator = gollm.NewFunction(
+	"calculator",
+	"A starlark calculator",
+	func(ctx *gollm.Context, args struct {
+		Expression string `description:"The expression to evaluate using starlark"`
+	}) (string, error) {
+		val, err := starlark.EvalOptions(&syntax.FileOptions{},
+			&starlark.Thread{Name: "main"},
+			"input",
+			args.Expression,
+			math.Module.Members)
+		if err != nil {
+			return "", err
+		}
+
+		return val.String(), nil
+	})