merge: integrate oss branch — budget gating, MCP back-channel, events, loopback-only bind

Key features from OSS branch:
- Budget gating: rolling per-provider spend caps with BUDGET_EXCEEDED task state
- Agent MCP back-channel: runners mint tokens; /mcp endpoint resolves them
- Chatbot MCP server at /chatbot/mcp (requires api_token in config)
- Event timeline: unified event stream replacing ad-hoc question/summary flows
- Loopback-only default bind (127.0.0.1:8484); external_bind_allowed=true to expose
- repository_url required on task creation (enforces traceability)
- Auto-checker: spawns verification task after each execution

Conflict resolutions:
- serve.go: keep cfg.Runners.XEnabled() conditional registration from main + agentRegistry from oss
- config.go: keep RunnersConfig from main + BudgetConfig/ExternalBindAllowed from oss
- server.go: handleStaticFiles (base-path rewrite) kept; deduplicated duplicate from both sides
- executor/claude.go: accepted oss deletion (ClaudeRunner replaced by ContainerRunner)
- container_test.go: added noopChannel + AgentChannel parameter to Run call

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

1 files changed, 135 insertions, 0 deletions

diff --git a/internal/executor/localtools.go b/internal/executor/localtools.go
new file mode 100644
index 0000000..48b862f
--- /dev/null
+++ b/internal/executor/localtools.go
@@ -0,0 +1,135 @@
+package executor
+
+import (
+	"context"
+	"encoding/json"
+	"errors"
+	"fmt"
+
+	"github.com/thepeterstone/claudomator/internal/llm"
+)
+
+// agentToolDefs returns the four agent back-channel tools as OpenAI
+// function-calling definitions, for runners that talk to an OpenAI-compatible
+// endpoint (LocalRunner). They mirror the MCP tools the container runners expose.
+func agentToolDefs() []llm.Tool {
+	strProp := func(desc string) map[string]any {
+		return map[string]any{"type": "string", "description": desc}
+	}
+	return []llm.Tool{
+		{Type: "function", Function: llm.ToolFunction{
+			Name:        "ask_user",
+			Description: "Ask the user a question when you genuinely need a decision to proceed. The task pauses until the user answers; do not call other tools after this.",
+			Parameters: map[string]any{
+				"type": "object",
+				"properties": map[string]any{
+					"question": strProp("the question to ask, phrased as a real question"),
+					"options":  map[string]any{"type": "array", "items": map[string]any{"type": "string"}, "description": "optional suggested answer choices"},
+				},
+				"required": []string{"question"},
+			},
+		}},
+		{Type: "function", Function: llm.ToolFunction{
+			Name:        "report_summary",
+			Description: "Record a concise 2-5 sentence summary of what you accomplished. Call this before finishing.",
+			Parameters: map[string]any{
+				"type":       "object",
+				"properties": map[string]any{"summary": strProp("the summary text")},
+				"required":   []string{"summary"},
+			},
+		}},
+		{Type: "function", Function: llm.ToolFunction{
+			Name:        "spawn_subtask",
+			Description: "Create a child task to be executed separately. Use this to break large work into focused pieces.",
+			Parameters: map[string]any{
+				"type": "object",
+				"properties": map[string]any{
+					"name":           strProp("short descriptive name for the subtask"),
+					"instructions":   strProp("complete instructions for the subtask agent"),
+					"model":          strProp("optional model override"),
+					"max_budget_usd": map[string]any{"type": "number", "description": "optional budget cap in USD"},
+				},
+				"required": []string{"name", "instructions"},
+			},
+		}},
+		{Type: "function", Function: llm.ToolFunction{
+			Name:        "record_progress",
+			Description: "Record a short progress note that appears in the task timeline.",
+			Parameters: map[string]any{
+				"type":       "object",
+				"properties": map[string]any{"message": strProp("a short progress note")},
+				"required":   []string{"message"},
+			},
+		}},
+	}
+}
+
+// dispatchAgentTool invokes one model-requested tool against the AgentChannel.
+// It returns the text to feed back to the model as the tool result, and a
+// blocked flag set when ask_user could not be answered in-session (the run must
+// stop and the task block).
+func dispatchAgentTool(ctx context.Context, ch AgentChannel, name, argsJSON string) (result string, blocked bool, err error) {
+	switch name {
+	case "ask_user":
+		var a struct {
+			Question string   `json:"question"`
+			Options  []string `json:"options"`
+		}
+		_ = json.Unmarshal([]byte(argsJSON), &a)
+		q := map[string]any{"text": a.Question}
+		if len(a.Options) > 0 {
+			q["options"] = a.Options
+		}
+		payload, _ := json.Marshal(q)
+		ans, askErr := ch.AskUser(ctx, string(payload))
+		if errors.Is(askErr, ErrAgentBlocked) {
+			return "", true, nil
+		}
+		if askErr != nil {
+			return "", false, askErr
+		}
+		return ans, false, nil
+
+	case "report_summary":
+		var a struct {
+			Summary string `json:"summary"`
+		}
+		_ = json.Unmarshal([]byte(argsJSON), &a)
+		if rsErr := ch.ReportSummary(ctx, a.Summary); rsErr != nil {
+			return "", false, rsErr
+		}
+		return "Summary recorded.", false, nil
+
+	case "spawn_subtask":
+		var a struct {
+			Name         string  `json:"name"`
+			Instructions string  `json:"instructions"`
+			Model        string  `json:"model"`
+			MaxBudgetUSD float64 `json:"max_budget_usd"`
+		}
+		_ = json.Unmarshal([]byte(argsJSON), &a)
+		id, ssErr := ch.SpawnSubtask(ctx, SubtaskSpec{
+			Name:         a.Name,
+			Instructions: a.Instructions,
+			Model:        a.Model,
+			MaxBudgetUSD: a.MaxBudgetUSD,
+		})
+		if ssErr != nil {
+			return "", false, ssErr
+		}
+		return "Created subtask " + id, false, nil
+
+	case "record_progress":
+		var a struct {
+			Message string `json:"message"`
+		}
+		_ = json.Unmarshal([]byte(argsJSON), &a)
+		if rpErr := ch.RecordProgress(ctx, a.Message); rpErr != nil {
+			return "", false, rpErr
+		}
+		return "Noted.", false, nil
+
+	default:
+		return "", false, fmt.Errorf("unknown tool %q", name)
+	}
+}