feat: harden Python workflow and structured planning

Replace the uv-specific skill with a portable Python skill and rewrite the default user system prompt as a concise, structured Markdown execution policy.

Normalize a single fenced structured-JSON response, retry invalid skill selection with think disabled, validate selected skill output, and make coding plans keep editing and verification in one task.

Document the file_edit patch payload, expose that contract in the tool catalog, and convert malformed patch input into a retryable tool failure instead of terminating the agent task.

Update configuration and runtime tests for the Python skill, exact tool catalog, structured-output retry, and implementation-and-verification planning guidance.

Verified with 82 Agent.Core tests and an agent-mcp Python scheduling scenario with 49 passing unittest cases.

1 files changed, 12 insertions, 1 deletions

diff --git a/tests/Agent.Core.Tests/PromptAndRuntimeTests.cs b/tests/Agent.Core.Tests/PromptAndRuntimeTests.cs
index fef5340..6dad528 100644
--- a/tests/Agent.Core.Tests/PromptAndRuntimeTests.cs
+++ b/tests/Agent.Core.Tests/PromptAndRuntimeTests.cs
@@ -284,6 +284,10 @@ public sealed class PromptAndRuntimeTests
         Assert.Equal(AgentTaskState.Completed, completed!.State);
         Assert.Equal([null, false], ollama.ThinkOverrides);
         Assert.All(ollama.Temperatures, value => Assert.Equal(0.7, value));
+        Assert.Contains(ollama.PlanningMessages[0], message =>
+            message.Role == "system" && message.Content.Contains("There is no file_write Tool", StringComparison.Ordinal));
+        Assert.Contains(ollama.PlanningMessages[0], message =>
+            message.Role == "system" && message.Content.Contains("one implementation-and-verification Task", StringComparison.Ordinal));
         Assert.Contains(ollama.PlanningMessages[1], message =>
             message.Role == "system" && message.Content.Contains("Do not call or execute any Tool", StringComparison.Ordinal));
     }
@@ -615,7 +619,14 @@ public sealed class PromptAndRuntimeTests
             Temperatures.Add(temperatureOverride);
             PlanningMessages.Add(messages);
             string content = ThinkOverrides.Count == 2 && succeedOnRetry
-                ? """{"tasks":[{"description":"analyze","dependencies":[],"budget_class":"4K","optimistic_tokens":10,"most_likely_tokens":20,"pessimistic_tokens":30,"requires_external_effect":false}]}"""
+                ? """
+                  Here is the requested plan.
+
+                  ```json
+                  {"tasks":[{"description":"analyze","dependencies":[],"budget_class":"4K","optimistic_tokens":10,"most_likely_tokens":20,"pessimistic_tokens":30,"requires_external_effect":false}]}
+                  ```
+                  The JSON above is the complete plan.
+                  """
                 : "not-json";
             return Task.FromResult(new OllamaResult(content, 1, 1));
         }