fix: avoid directive hits inside URLs and add tests

src/auto-reply/reply.directive.test.ts

@@ -0,0 +1,29 @@
+import { describe, expect, it } from "vitest";
+import { extractVerboseDirective, extractThinkDirective } from "./reply.js";
+
+describe("directive parsing", () => {
+  it("ignores verbose directive inside URL", () => {
+    const body = "https://x.com/verioussmith/status/1997066835133669687";
+    const res = extractVerboseDirective(body);
+    expect(res.hasDirective).toBe(false);
+    expect(res.cleaned).toBe(body);
+  });
+
+  it("ignores think directive inside URL", () => {
+    const body = "see https://example.com/path/thinkstuff";
+    const res = extractThinkDirective(body);
+    expect(res.hasDirective).toBe(false);
+  });
+
+  it("matches verbose with leading space", () => {
+    const res = extractVerboseDirective(" please /verbose on now");
+    expect(res.hasDirective).toBe(true);
+    expect(res.verboseLevel).toBe("on");
+  });
+
+  it("matches think at start of line", () => {
+    const res = extractThinkDirective("/think:high run slow");
+    expect(res.hasDirective).toBe(true);
+    expect(res.thinkLevel).toBe("high");
+  });
+});

src/auto-reply/reply.ts

@@ -36,15 +36,18 @@ export type { GetReplyOptions, ReplyPayload } from "./types.js";
 const ABORT_TRIGGERS = new Set(["stop", "esc", "abort", "wait", "exit"]);
 const ABORT_MEMORY = new Map<string, boolean>();
 
-function extractThinkDirective(body?: string): {
+export function extractThinkDirective(body?: string): {
   cleaned: string;
   thinkLevel?: ThinkLevel;
   rawLevel?: string;
   hasDirective: boolean;
 } {
   if (!body) return { cleaned: "", hasDirective: false };
-  // Match the longest keyword first to avoid partial captures (e.g. "/think:high")
-  const match = body.match(/\/(?:thinking|think|t)\s*:?\s*([a-zA-Z-]+)\b/i);
+  // Match the longest keyword first to avoid partial captures (e.g. "/think:high").
+  // Require start of string or whitespace before "/" to avoid catching URLs.
+  const match = body.match(
+    /(?:^|\s)\/(?:thinking|think|t)\s*:?\s*([a-zA-Z-]+)\b/i,
+  );
   const thinkLevel = normalizeThinkLevel(match?.[1]);
   const cleaned = match
     ? body.replace(match[0], "").replace(/\s+/g, " ").trim()
@@ -57,14 +60,15 @@ function extractThinkDirective(body?: string): {
   };
 }
 
-function extractVerboseDirective(body?: string): {
+export function extractVerboseDirective(body?: string): {
   cleaned: string;
   verboseLevel?: VerboseLevel;
   rawLevel?: string;
   hasDirective: boolean;
 } {
   if (!body) return { cleaned: "", hasDirective: false };
-  const match = body.match(/\/(?:verbose|v)\s*:?\s*([a-zA-Z-]+)\b/i);
+  // Require start or whitespace before "/verbose" to avoid matching URLs like /verioussmith.
+  const match = body.match(/(?:^|\s)\/(?:verbose|v)\s*:?\s*([a-zA-Z-]+)\b/i);
   const verboseLevel = normalizeVerboseLevel(match?.[1]);
   const cleaned = match
     ? body.replace(match[0], "").replace(/\s+/g, " ").trim()