diff --git a/.changeset/document-exec-usage.md b/.changeset/document-exec-usage.md
new file mode 100644
index 00000000..c9101bdc
--- /dev/null
+++ b/.changeset/document-exec-usage.md
@@ -0,0 +1,9 @@
+---
+"create-expert": patch
+---
+
+Document why exec is used in expert-tester
+
+Added explanation that exec is used instead of delegation because:
+- Tests need to run Experts as black-box (same as end-users)
+- CLI execution ensures realistic test conditions
diff --git a/apps/create-expert/src/lib/create-expert-toml.ts b/apps/create-expert/src/lib/create-expert-toml.ts
index db4e42f9..a5fb3190 100644
--- a/apps/create-expert/src/lib/create-expert-toml.ts
+++ b/apps/create-expert/src/lib/create-expert-toml.ts
@@ -375,6 +375,10 @@ From the stage manager:
 ## Testing Process
 
 ### 1. Execute Tests
+
+NOTE: We use \`exec\` instead of delegation because we need to test the Expert as a black-box,
+exactly as end-users would run it via the CLI. This ensures realistic test conditions.
+
 For each test case, run:
 \`\`\`bash
 npx -y perstack run expert-name "test query" --workspace . --filter completeRun