feat: AI Agent Sandbox MCP Server v1を実装

Specifications、Design、ImplementationPlanの優先順位を整理し、単一Task実行モデル、typed option優先、設定可能なresource上限、local Ollamaを必須とするRelease Gateを文書化。

.NET 10/C# 14の5 production project、3 test project、E2E Clientを構築。typed YAML、en/ja locale、fixed/user prompt、stdioおよびloopback限定Streamable HTTP、agent_run/agent_status、Session、Ollama text/tool/Vision、TaskGraphとtoken accountingを実装。

Windows sandboxとしてfixed AppContainer、canonical/reparse検査、能力別ACL、path+SHA-256 executable approval、child-process restriction、Job Object、task journal、atomic file transaction、crash recoveryを実装。WFPのIPv4/IPv6 default denyとexact allow、dynamic session、temporary loopback exemption、管理者/BFE診断を追加。

Security/Recovery test、MCP/Ollama integration test、stdio/HTTP Operational E2E、Release Bundle生成・checksum・manifest・fresh extraction検査を追加。Requirement TraceabilityとPhase 0〜11の進捗、初回セットアップ可能なREADMEを整備。

Debug/Release/publish/objをrepository rootのbuild/へ集約し、旧project-local bin/objへの依存を除去。artifacts/とbuild/をGit管理外に設定し、clean checkoutでもsolution buildからE2E Clientまで生成可能にした。

3 files changed, 194 insertions, 0 deletions

diff --git a/src/Agent.Core/Runtime/AgentRuntime.cs b/src/Agent.Core/Runtime/AgentRuntime.cs
new file mode 100644
index 0000000..c2fa568
--- /dev/null
+++ b/src/Agent.Core/Runtime/AgentRuntime.cs
@@ -0,0 +1,165 @@
+using Agent.Core.Ollama;
+using Agent.Core.Prompts;
+using Agent.Core.Sessions;
+using Agent.Core.Tasks;
+using Agent.Core.Tools;
+
+namespace Agent.Core.Runtime;
+
+public sealed class AgentRuntime(TaskRegistry registry, IOllamaClient ollama, AgentRuntimeSettings settings, InternalToolDispatcher? toolDispatcher = null, IAgentRuntimeObserver? observer = null)
+{
+    private readonly InternalToolDispatcher _tools = toolDispatcher ?? new(
+        new DenyByDefaultPolicy(InternalToolCatalog.All.Select(static item => item.Name).ToHashSet(StringComparer.Ordinal)),
+        new FakeInternalToolBackend());
+    private readonly IAgentRuntimeObserver _observer = observer ?? NullAgentRuntimeObserver.Instance;
+
+    public async Task ExecuteAsync(Guid taskId, CancellationToken cancellationToken)
+    {
+        if (!registry.TryGet(taskId, null, out TaskSnapshot? task) || task is null) return;
+        if (!registry.Sessions.TryGet(task.SessionId, out AgentSession? session) || session is null)
+        {
+            registry.Transition(taskId, AgentTaskState.Failed, result: "Session not found.");
+            return;
+        }
+
+        try
+        {
+            await _tools.BeginTaskAsync(taskId, cancellationToken).ConfigureAwait(false);
+            if (task.Images.Count > 0 && await ollama.SupportsVisionAsync(cancellationToken).ConfigureAwait(false) is false)
+                throw new InvalidOperationException(settings.VisionUnsupportedMessage);
+            var messages = PromptComposer.Compose(new PromptContext(
+                FixedPrompt.Load(),
+                settings.UserSystemPrompt,
+                settings.Locale,
+                settings.Skills,
+                settings.SandboxCapabilitySummary,
+                "Current task: " + task.Request,
+                string.Empty,
+                session.Conversation,
+                task.Request,
+                task.Images)).ToList();
+            var guard = new ToolUseGuard(settings.SameFailureLimit);
+            bool requiresExternalEffect = RequiresExternalEffect(task.Request);
+            bool externalEffectEvidence = false;
+            var unresolvedToolFailures = new Dictionary<string, string>(StringComparer.Ordinal);
+            IReadOnlyDictionary<string, int> requiredToolCalls = RequiredToolCalls(task.Request);
+            var successfulToolCalls = new Dictionary<string, int>(StringComparer.Ordinal);
+            string? finalContent = null;
+            for (int iteration = 0; iteration < settings.MaxIterations; iteration++)
+            {
+                _observer.LlmRequest(taskId, iteration, messages);
+                OllamaResult result = await ollama.ChatWithToolsAsync(messages, task.Temperature, InternalToolCatalog.All, cancellationToken).ConfigureAwait(false);
+                _observer.LlmResponse(taskId, iteration, result);
+                registry.AddUsage(taskId, result.InputTokens, result.OutputTokens);
+                IReadOnlyList<OllamaToolCall> calls = result.ToolCalls ?? [];
+                if (calls.Count > 0)
+                {
+                    messages.Add(new OllamaMessage("assistant", result.Content, ToolCalls: calls));
+                    string? failureFingerprint = null;
+                    foreach (OllamaToolCall call in calls)
+                    {
+                        ToolExecutionResult toolResult;
+                        try
+                        {
+                            toolResult = await _tools.DispatchAsync(taskId, call.Function.Name, call.Function.Arguments, cancellationToken).ConfigureAwait(false);
+                        }
+                        catch (Exception ex) when (ex is ToolValidationException or ArgumentException or UnauthorizedAccessException)
+                        {
+                            toolResult = new(false, ex.Message, false, $"{call.Function.Name}:{ex.GetType().Name}:{ex.Message}");
+                        }
+                        externalEffectEvidence |= toolResult.Success && toolResult.ExternalEffect;
+                        if (toolResult.Success) unresolvedToolFailures.Remove(call.Function.Name);
+                        else unresolvedToolFailures[call.Function.Name] = toolResult.FailureFingerprint ?? toolResult.Output;
+                        if (toolResult.Success) successfulToolCalls[call.Function.Name] = successfulToolCalls.GetValueOrDefault(call.Function.Name) + 1;
+                        _observer.ToolResult(taskId, call.Function.Name, toolResult);
+                        failureFingerprint = toolResult.FailureFingerprint ?? failureFingerprint;
+                        messages.Add(new OllamaMessage("tool", toolResult.Output, ToolName: call.Function.Name));
+                    }
+                    ToolUseGuardResult callGuard = guard.Evaluate(requiresExternalEffect, true, externalEffectEvidence, failureFingerprint);
+                    if (callGuard.Decision == ToolUseGuardDecision.Fail) throw new InvalidOperationException(callGuard.Observation);
+                    continue;
+                }
+
+                if (unresolvedToolFailures.Count > 0)
+                {
+                    messages.Add(new OllamaMessage("system", "Completion is not allowed while tool failures remain unresolved. Retry or correct: " +
+                        string.Join("; ", unresolvedToolFailures.Select(static item => $"{item.Key}={item.Value}"))));
+                    continue;
+                }
+                string[] missingCalls = requiredToolCalls
+                    .Where(requirement => successfulToolCalls.GetValueOrDefault(requirement.Key) < requirement.Value)
+                    .Select(requirement => $"{requirement.Key}: {successfulToolCalls.GetValueOrDefault(requirement.Key)}/{requirement.Value}")
+                    .ToArray();
+                if (missingCalls.Length > 0)
+                {
+                    messages.Add(new OllamaMessage("system", "Completion is not allowed because the request explicitly requires successful tool calls that are still missing: " + string.Join("; ", missingCalls)));
+                    continue;
+                }
+                ToolUseGuardResult completion = guard.Evaluate(requiresExternalEffect, false, externalEffectEvidence);
+                if (completion.Decision == ToolUseGuardDecision.Retry)
+                {
+                    messages.Add(new OllamaMessage("system", completion.Observation));
+                    continue;
+                }
+                if (completion.Decision == ToolUseGuardDecision.Fail) throw new InvalidOperationException(completion.Observation);
+                finalContent = result.Content;
+                break;
+            }
+            if (finalContent is null) throw new InvalidOperationException($"Agent exceeded the iteration limit of {settings.MaxIterations}.");
+            string? cleanupFailure = await EndTaskSafelyAsync(taskId, true, CancellationToken.None).ConfigureAwait(false);
+            if (cleanupFailure is not null) throw new InvalidOperationException(cleanupFailure);
+            session.Append(new ConversationMessage("user", task.Request));
+            session.Append(new ConversationMessage("assistant", finalContent));
+            registry.Transition(taskId, AgentTaskState.Completed, result: finalContent);
+        }
+        catch (OperationCanceledException) when (cancellationToken.IsCancellationRequested)
+        {
+            string? cleanupFailure = await EndTaskSafelyAsync(taskId, false, CancellationToken.None).ConfigureAwait(false);
+            string result = cleanupFailure is null ? "Cancelled." : $"Cancelled. {cleanupFailure}";
+            registry.Transition(taskId, AgentTaskState.Cancelled, result: result);
+        }
+        catch (Exception ex) when (ex is not (OutOfMemoryException or StackOverflowException or AccessViolationException))
+        {
+            string? cleanupFailure = await EndTaskSafelyAsync(taskId, false, CancellationToken.None).ConfigureAwait(false);
+            string result = cleanupFailure is null ? ex.Message : $"{ex.Message} {cleanupFailure}";
+            registry.Transition(taskId, AgentTaskState.Failed, result: result);
+        }
+    }
+
+    private async Task<string?> EndTaskSafelyAsync(Guid taskId, bool success, CancellationToken cancellationToken)
+    {
+        try
+        {
+            await _tools.EndTaskAsync(taskId, success, cancellationToken).ConfigureAwait(false);
+            return null;
+        }
+        catch (Exception ex) when (ex is not (OutOfMemoryException or StackOverflowException or AccessViolationException))
+        {
+            return $"Task resource cleanup failed: {ex.Message}";
+        }
+    }
+
+    private static bool RequiresExternalEffect(string request)
+    {
+        string normalized = request.ToLowerInvariant();
+        string[] indicators = ["edit", "modify", "change", "fix", "write", "delete", "create", "build", "test", "run", "修正", "変更", "編集", "作成", "削除", "ビルド", "テスト", "実行"];
+        return indicators.Any(normalized.Contains);
+    }
+
+    private static IReadOnlyDictionary<string, int> RequiredToolCalls(string request)
+    {
+        var result = new Dictionary<string, int>(StringComparer.Ordinal);
+        foreach (InternalToolDefinition tool in InternalToolCatalog.All)
+        {
+            int count = 0;
+            int offset = 0;
+            while ((offset = request.IndexOf(tool.Name, offset, StringComparison.OrdinalIgnoreCase)) >= 0)
+            {
+                count++;
+                offset += tool.Name.Length;
+            }
+            if (count > 0) result[tool.Name] = count;
+        }
+        return result;
+    }
+}
diff --git a/src/Agent.Core/Runtime/AgentRuntimeSettings.cs b/src/Agent.Core/Runtime/AgentRuntimeSettings.cs
new file mode 100644
index 0000000..9bc04b9
--- /dev/null
+++ b/src/Agent.Core/Runtime/AgentRuntimeSettings.cs
@@ -0,0 +1,10 @@
+namespace Agent.Core.Runtime;
+
+public sealed record AgentRuntimeSettings(
+    string UserSystemPrompt,
+    string Locale,
+    IReadOnlyList<string> Skills,
+    string SandboxCapabilitySummary,
+    int MaxIterations = 128,
+    int SameFailureLimit = 5,
+    string VisionUnsupportedMessage = "The selected model does not support image input.");
diff --git a/src/Agent.Core/Runtime/IAgentRuntimeObserver.cs b/src/Agent.Core/Runtime/IAgentRuntimeObserver.cs
new file mode 100644
index 0000000..e4e26e3
--- /dev/null
+++ b/src/Agent.Core/Runtime/IAgentRuntimeObserver.cs
@@ -0,0 +1,19 @@
+using Agent.Core.Ollama;
+using Agent.Core.Tools;
+
+namespace Agent.Core.Runtime;
+
+public interface IAgentRuntimeObserver
+{
+    void LlmRequest(Guid taskId, int iteration, IReadOnlyList<OllamaMessage> messages);
+    void LlmResponse(Guid taskId, int iteration, OllamaResult result);
+    void ToolResult(Guid taskId, string toolName, ToolExecutionResult result);
+}
+
+public sealed class NullAgentRuntimeObserver : IAgentRuntimeObserver
+{
+    public static NullAgentRuntimeObserver Instance { get; } = new();
+    public void LlmRequest(Guid taskId, int iteration, IReadOnlyList<OllamaMessage> messages) { }
+    public void LlmResponse(Guid taskId, int iteration, OllamaResult result) { }
+    public void ToolResult(Guid taskId, string toolName, ToolExecutionResult result) { }
+}