chore: upgrade claude-agent-sdk to 0.2.11 and improve task judge feedback

package-lock.json

@@ -16,7 +16,7 @@
         "packages/website"
       ],
       "dependencies": {
-        "@anthropic-ai/claude-agent-sdk": "^0.1.75"
+        "@anthropic-ai/claude-agent-sdk": "^0.2.11"
       },
       "devDependencies": {
         "patch-package": "^8.0.1",
@@ -100,9 +100,9 @@
       }
     },
     "node_modules/@anthropic-ai/claude-agent-sdk": {
-      "version": "0.1.75",
+      "version": "0.2.11",
-      "resolved": "https://registry.npmjs.org/@anthropic-ai/claude-agent-sdk/-/claude-agent-sdk-0.1.75.tgz",
+      "resolved": "https://registry.npmjs.org/@anthropic-ai/claude-agent-sdk/-/claude-agent-sdk-0.2.11.tgz",
-      "integrity": "sha512-8iYosqTq98pm6Z1IJ0OY50mpkSZ7fdSO8cJJjFHXiKHCwfmiPZ05vLYUed2p2an9GPLpXtJhcGDfrgseenrgPw==",
+      "integrity": "sha512-50q4vfh57HYTUrwukULp3gvSaOZfRF5zukxhWvW6mmUHuD8+Xwhqn59sZhoDz9ZUw6Um+1lUCgWpJw5/TTMn5w==",
       "license": "SEE LICENSE IN README.md",
       "engines": {
         "node": ">=18.0.0"
@@ -118,7 +118,7 @@
         "@img/sharp-win32-x64": "^0.33.5"
       },
       "peerDependencies": {
-        "zod": "^3.24.1 || ^4.0.0"
+        "zod": "^4.0.0"
       }
     },
     "node_modules/@babel/code-frame": {
@@ -3292,16 +3292,6 @@
         "node": ">=10"
       }
     },
-    "node_modules/@expo/eas-build-job/node_modules/zod": {
-      "version": "4.1.12",
-      "resolved": "https://registry.npmjs.org/zod/-/zod-4.1.12.tgz",
-      "integrity": "sha512-JInaHOamG8pt5+Ey8kGmdcAcg3OL9reK8ltczgHTAwNhMys/6ThXHityHxVV2p3fkw/c+MAvBHFVYHFZDmjMCQ==",
-      "dev": true,
-      "license": "MIT",
-      "funding": {
-        "url": "https://github.com/sponsors/colinhacks"
-      }
-    },
     "node_modules/@expo/eas-json": {
       "version": "16.23.0",
       "resolved": "https://registry.npmjs.org/@expo/eas-json/-/eas-json-16.23.0.tgz",
@@ -3500,16 +3490,6 @@
       "dev": true,
       "license": "ISC"
     },
-    "node_modules/@expo/eas-json/node_modules/zod": {
-      "version": "4.1.12",
-      "resolved": "https://registry.npmjs.org/zod/-/zod-4.1.12.tgz",
-      "integrity": "sha512-JInaHOamG8pt5+Ey8kGmdcAcg3OL9reK8ltczgHTAwNhMys/6ThXHityHxVV2p3fkw/c+MAvBHFVYHFZDmjMCQ==",
-      "dev": true,
-      "license": "MIT",
-      "funding": {
-        "url": "https://github.com/sponsors/colinhacks"
-      }
-    },
     "node_modules/@expo/env": {
       "version": "2.0.8",
       "resolved": "https://registry.npmjs.org/@expo/env/-/env-2.0.8.tgz",
@@ -3695,6 +3675,15 @@
         }
       }
     },
+    "node_modules/@expo/mcp-tunnel/node_modules/zod": {
+      "version": "3.25.76",
+      "resolved": "https://registry.npmjs.org/zod/-/zod-3.25.76.tgz",
+      "integrity": "sha512-gzUt/qt81nXsFGKIFcC3YnfEAx5NkunCfnDlvuBSSFS02bcXu4Lmea0AFIUwbLWxWPx3d9p8S5QoaujKcNQxcQ==",
+      "license": "MIT",
+      "funding": {
+        "url": "https://github.com/sponsors/colinhacks"
+      }
+    },
     "node_modules/@expo/metro": {
       "version": "54.1.0",
       "resolved": "https://registry.npmjs.org/@expo/metro/-/metro-54.1.0.tgz",
@@ -4581,16 +4570,6 @@
         "node": ">=10"
       }
     },
-    "node_modules/@expo/steps/node_modules/zod": {
-      "version": "4.1.12",
-      "resolved": "https://registry.npmjs.org/zod/-/zod-4.1.12.tgz",
-      "integrity": "sha512-JInaHOamG8pt5+Ey8kGmdcAcg3OL9reK8ltczgHTAwNhMys/6ThXHityHxVV2p3fkw/c+MAvBHFVYHFZDmjMCQ==",
-      "dev": true,
-      "license": "MIT",
-      "funding": {
-        "url": "https://github.com/sponsors/colinhacks"
-      }
-    },
     "node_modules/@expo/sudo-prompt": {
       "version": "9.3.2",
       "resolved": "https://registry.npmjs.org/@expo/sudo-prompt/-/sudo-prompt-9.3.2.tgz",
@@ -6008,6 +5987,17 @@
         "node": ">= 0.6"
       }
     },
+    "node_modules/@modelcontextprotocol/sdk/node_modules/zod": {
+      "version": "3.25.76",
+      "resolved": "https://registry.npmjs.org/zod/-/zod-3.25.76.tgz",
+      "integrity": "sha512-gzUt/qt81nXsFGKIFcC3YnfEAx5NkunCfnDlvuBSSFS02bcXu4Lmea0AFIUwbLWxWPx3d9p8S5QoaujKcNQxcQ==",
+      "license": "MIT",
+      "optional": true,
+      "peer": true,
+      "funding": {
+        "url": "https://github.com/sponsors/colinhacks"
+      }
+    },
     "node_modules/@napi-rs/wasm-runtime": {
       "version": "0.2.12",
       "resolved": "https://registry.npmjs.org/@napi-rs/wasm-runtime/-/wasm-runtime-0.2.12.tgz",
@@ -8185,6 +8175,15 @@
         "node": ">= 12"
       }
     },
+    "node_modules/@tanstack/router-generator/node_modules/zod": {
+      "version": "3.25.76",
+      "resolved": "https://registry.npmjs.org/zod/-/zod-3.25.76.tgz",
+      "integrity": "sha512-gzUt/qt81nXsFGKIFcC3YnfEAx5NkunCfnDlvuBSSFS02bcXu4Lmea0AFIUwbLWxWPx3d9p8S5QoaujKcNQxcQ==",
+      "license": "MIT",
+      "funding": {
+        "url": "https://github.com/sponsors/colinhacks"
+      }
+    },
     "node_modules/@tanstack/router-plugin": {
       "version": "1.149.3",
       "resolved": "https://registry.npmjs.org/@tanstack/router-plugin/-/router-plugin-1.149.3.tgz",
@@ -8238,6 +8237,15 @@
         }
       }
     },
+    "node_modules/@tanstack/router-plugin/node_modules/zod": {
+      "version": "3.25.76",
+      "resolved": "https://registry.npmjs.org/zod/-/zod-3.25.76.tgz",
+      "integrity": "sha512-gzUt/qt81nXsFGKIFcC3YnfEAx5NkunCfnDlvuBSSFS02bcXu4Lmea0AFIUwbLWxWPx3d9p8S5QoaujKcNQxcQ==",
+      "license": "MIT",
+      "funding": {
+        "url": "https://github.com/sponsors/colinhacks"
+      }
+    },
     "node_modules/@tanstack/router-utils": {
       "version": "1.143.11",
       "resolved": "https://registry.npmjs.org/@tanstack/router-utils/-/router-utils-1.143.11.tgz",
@@ -8332,6 +8340,15 @@
         "vite": ">=7.0.0"
       }
     },
+    "node_modules/@tanstack/start-plugin-core/node_modules/zod": {
+      "version": "3.25.76",
+      "resolved": "https://registry.npmjs.org/zod/-/zod-3.25.76.tgz",
+      "integrity": "sha512-gzUt/qt81nXsFGKIFcC3YnfEAx5NkunCfnDlvuBSSFS02bcXu4Lmea0AFIUwbLWxWPx3d9p8S5QoaujKcNQxcQ==",
+      "license": "MIT",
+      "funding": {
+        "url": "https://github.com/sponsors/colinhacks"
+      }
+    },
     "node_modules/@tanstack/start-server-core": {
       "version": "1.149.3",
       "resolved": "https://registry.npmjs.org/@tanstack/start-server-core/-/start-server-core-1.149.3.tgz",
@@ -12316,16 +12333,6 @@
       "dev": true,
       "license": "ISC"
     },
-    "node_modules/eas-cli/node_modules/zod": {
-      "version": "4.1.12",
-      "resolved": "https://registry.npmjs.org/zod/-/zod-4.1.12.tgz",
-      "integrity": "sha512-JInaHOamG8pt5+Ey8kGmdcAcg3OL9reK8ltczgHTAwNhMys/6ThXHityHxVV2p3fkw/c+MAvBHFVYHFZDmjMCQ==",
-      "dev": true,
-      "license": "MIT",
-      "funding": {
-        "url": "https://github.com/sponsors/colinhacks"
-      }
-    },
     "node_modules/eastasianwidth": {
       "version": "0.2.0",
       "resolved": "https://registry.npmjs.org/eastasianwidth/-/eastasianwidth-0.2.0.tgz",
@@ -17751,6 +17758,15 @@
         }
       }
     },
+    "node_modules/miniflare/node_modules/zod": {
+      "version": "3.25.76",
+      "resolved": "https://registry.npmjs.org/zod/-/zod-3.25.76.tgz",
+      "integrity": "sha512-gzUt/qt81nXsFGKIFcC3YnfEAx5NkunCfnDlvuBSSFS02bcXu4Lmea0AFIUwbLWxWPx3d9p8S5QoaujKcNQxcQ==",
+      "license": "MIT",
+      "funding": {
+        "url": "https://github.com/sponsors/colinhacks"
+      }
+    },
     "node_modules/minimatch": {
       "version": "9.0.5",
       "resolved": "https://registry.npmjs.org/minimatch/-/minimatch-9.0.5.tgz",
@@ -24514,21 +24530,21 @@
       }
     },
     "node_modules/zod": {
-      "version": "3.25.76",
+      "version": "4.3.5",
-      "resolved": "https://registry.npmjs.org/zod/-/zod-3.25.76.tgz",
+      "resolved": "https://registry.npmjs.org/zod/-/zod-4.3.5.tgz",
-      "integrity": "sha512-gzUt/qt81nXsFGKIFcC3YnfEAx5NkunCfnDlvuBSSFS02bcXu4Lmea0AFIUwbLWxWPx3d9p8S5QoaujKcNQxcQ==",
+      "integrity": "sha512-k7Nwx6vuWx1IJ9Bjuf4Zt1PEllcwe7cls3VNzm4CQ1/hgtFUK2bRNG3rvnpPUhFjmqJKAKtjV576KnUkHocg/g==",
       "license": "MIT",
       "funding": {
         "url": "https://github.com/sponsors/colinhacks"
       }
     },
     "node_modules/zod-to-json-schema": {
-      "version": "3.24.6",
+      "version": "3.25.1",
-      "resolved": "https://registry.npmjs.org/zod-to-json-schema/-/zod-to-json-schema-3.24.6.tgz",
+      "resolved": "https://registry.npmjs.org/zod-to-json-schema/-/zod-to-json-schema-3.25.1.tgz",
-      "integrity": "sha512-h/z3PKvcTcTetyjl1fkj79MHNEjm+HpD6NXheWjzOekY7kV+lwDYnHw+ivHkijnCSMz1yJaWBD9vu/Fcmk+vEg==",
+      "integrity": "sha512-pM/SU9d3YAggzi6MtR4h7ruuQlqKtad8e9S0fmxcMi+ueAK5Korys/aWcV9LIIHTVbj01NdzxcnXSN+O74ZIVA==",
       "license": "ISC",
       "peerDependencies": {
-        "zod": "^3.24.1"
+        "zod": "^3.25 || ^4"
       }
     },
     "node_modules/zustand": {
@@ -24650,6 +24666,15 @@
         "react-native": "*"
       }
     },
+    "packages/app/node_modules/zod": {
+      "version": "3.25.76",
+      "resolved": "https://registry.npmjs.org/zod/-/zod-3.25.76.tgz",
+      "integrity": "sha512-gzUt/qt81nXsFGKIFcC3YnfEAx5NkunCfnDlvuBSSFS02bcXu4Lmea0AFIUwbLWxWPx3d9p8S5QoaujKcNQxcQ==",
+      "license": "MIT",
+      "funding": {
+        "url": "https://github.com/sponsors/colinhacks"
+      }
+    },
     "packages/relay": {
       "name": "@paseo/relay",
       "version": "0.1.0",
@@ -24972,6 +24997,15 @@
         "node": ">= 0.6"
       }
     },
+    "packages/server/node_modules/zod": {
+      "version": "3.25.76",
+      "resolved": "https://registry.npmjs.org/zod/-/zod-3.25.76.tgz",
+      "integrity": "sha512-gzUt/qt81nXsFGKIFcC3YnfEAx5NkunCfnDlvuBSSFS02bcXu4Lmea0AFIUwbLWxWPx3d9p8S5QoaujKcNQxcQ==",
+      "license": "MIT",
+      "funding": {
+        "url": "https://github.com/sponsors/colinhacks"
+      }
+    },
     "packages/website": {
       "name": "@paseo/website",
       "version": "1.0.0",

package.json

@@ -42,6 +42,6 @@
     "lightningcss": "1.30.1"
   },
   "dependencies": {
-    "@anthropic-ai/claude-agent-sdk": "^0.1.75"
+    "@anthropic-ai/claude-agent-sdk": "^0.2.11"
   }
 }

packages/server/src/server/logger.ts

@@ -16,7 +16,7 @@ export function resolveLogConfig(
   const envFormat = process.env.PASEO_LOG_FORMAT as LogFormat | undefined;
 
   const level: LogLevel =
-    envLevel ?? persistedConfig?.log?.level ?? "info";
+    envLevel ?? persistedConfig?.log?.level ?? "debug";
   const format: LogFormat =
     envFormat ?? persistedConfig?.log?.format ?? "pretty";
 

packages/server/src/tasks/cli.ts

@@ -971,9 +971,28 @@ The ONLY question: Does the acceptance criterion pass when verified? Yes or No.
 
 If an agent left notes explaining why something couldn't be done, IGNORE THE EXPLANATION. Just check: is the criterion met?
 
+## REQUIRED: Provide Feedback to the Worker
+
+After verifying, you MUST leave feedback for the worker by adding a note. The worker will read
+this note in the next iteration - it's their ONLY way to know what went wrong and what's missing.
+
+\`\`\`
+task note ${task.id} "JUDGE: [verdict] - [specific feedback for the worker]"
+\`\`\`
+
+Your feedback should tell the worker exactly what to fix:
+- Which specific tests failed and with what error messages
+- Which files or commands still need attention
+- What's missing from the acceptance criteria
+
+BAD: "tests failed" (useless - worker doesn't know which ones)
+GOOD: "3 tests fail: codex-mcp-agent.test.ts 'Session not found', opencode-agent.test.ts timeout after 60s, tool-calls.e2e.test.ts shell assertion"
+
+The worker cannot see your output - only the notes you leave.
+
 ## Output Format - CRITICAL
 
-You MUST output one of these XML tags at the END of your response. This is REQUIRED for the system to parse your verdict:
+You MUST output one of these XML tags at the END of your response:
 
 If ALL criteria pass:
 <VERDICT>DONE</VERDICT>
@@ -981,10 +1000,7 @@ If ALL criteria pass:
 If ANY criterion fails:
 <VERDICT>NOT_DONE</VERDICT>
 
-If you do not include this exact XML tag, your verdict will not be recorded and the task will retry.
+If you do not include this exact XML tag, your verdict will not be recorded.
-
-Then add a note to the task with your findings:
-\`task note ${task.id} "JUDGE: [DONE/NOT_DONE] - details..."\`
 `;
 }