refactor: replace orchestrator/verifier chain with direct LiteLLM calls

Drop the three-layer Claude subprocess orchestration (local model →
Claude verifier → cloud escalation). Skills now call LiteLLM directly
and return plain text to Claude Code, which decides what to do with it.

- Delete executor, orchestrator, verifier, result, attempts packages
- Simplify LiteLLMExecutor: Run(Request)→Result becomes Complete(model,sys,user)→(string,int64,error)
- Replace ExecutorFn with CompleteFunc in all 6 skill configs
- Rewrite all skill handlers to call Complete and return {"text","model","duration_ms"}
- Simplify config/models: remove Verifier/LlamaSwapURL, add ModelFor
- Bump version to v0.5.0

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

internal/skills/trainer/handlers.go

@@ -7,7 +7,6 @@ import (
 	"fmt"
 	"time"
 
-	iexec "github.com/mathiasbq/supervisor/internal/exec"
 	"github.com/mathiasbq/supervisor/internal/session"
 )
 
@@ -28,7 +27,7 @@ func (s *Skill) Handle(ctx context.Context, tool string, args json.RawMessage) (
 	if a.SessionID == "" {
 		return nil, fmt.Errorf("session_id is required")
 	}
-	if s.cfg.ExecutorFn == nil {
+	if s.cfg.CompleteFunc == nil {
 		return nil, fmt.Errorf("no executor configured")
 	}
 
@@ -42,53 +41,47 @@ func (s *Skill) Handle(ctx context.Context, tool string, args json.RawMessage) (
 		return nil, fmt.Errorf("read session log: %w", err)
 	}
 
-	// ── Step 1: Reader agent ─────────────────────────────────────────────────
+	// ── Step 1: Reader ────────────────────────────────────────────────────────
 	history := session.FormatHistory(entries, "")
 	readerTask := fmt.Sprintf(
 		"role: reader\nsession_id: %s\nbrain_dir: %s\n\n%s",
 		a.SessionID, s.cfg.BrainDir, history,
 	)
-	readerResult, err := s.cfg.ExecutorFn(ctx, iexec.Request{
-		SkillPrompt: s.cfg.ReaderPrompt,
-		TaskPrompt:  readerTask,
-		Model:       model,
-		Tools:       "Read",
-	})
+	readerText, _, err := s.cfg.CompleteFunc(ctx, model, s.cfg.ReaderPrompt, readerTask)
 	if err != nil {
-		return nil, fmt.Errorf("reader agent: %w", err)
+		return nil, fmt.Errorf("reader: %w", err)
 	}
 
-	// ── Step 2: Writer agent (receives reader candidates) ────────────────────
+	// ── Step 2: Writer (receives reader output) ───────────────────────────────
 	t0 := time.Now()
 	writerTask := fmt.Sprintf(
-		"role: writer\nsession_id: %s\nbrain_dir: %s\n\nreader_summary: %s\nreader_candidates:\n%s",
-		a.SessionID, s.cfg.BrainDir, readerResult.Message, readerResult.RunnerOutput,
+		"role: writer\nsession_id: %s\nbrain_dir: %s\n\nreader_analysis:\n%s",
+		a.SessionID, s.cfg.BrainDir, readerText,
 	)
-	writerResult, err := s.cfg.ExecutorFn(ctx, iexec.Request{
-		SkillPrompt: s.cfg.WriterPrompt,
-		TaskPrompt:  writerTask,
-		Model:       model,
-		Tools:       "Read,Write",
-	})
+	writerText, dur, err := s.cfg.CompleteFunc(ctx, model, s.cfg.WriterPrompt, writerTask)
 	if err != nil {
-		return nil, fmt.Errorf("writer agent: %w", err)
+		return nil, fmt.Errorf("writer: %w", err)
 	}
 
+	msg := writerText
+	if len(msg) > 200 {
+		msg = msg[:200]
+	}
 	_ = session.Append(s.cfg.SessionsDir, a.SessionID, session.Entry{
 		SessionID:   a.SessionID,
 		Timestamp:   time.Now(),
 		Skill:       "trainer",
 		Phase:       "trainer",
-		Attempts:    session.AttemptsFrom(writerResult.Attempts),
-		FinalStatus: writerResult.Status,
-		ModelUsed:   writerResult.ModelUsed,
+		FinalStatus: "ok",
+		ModelUsed:   model,
 		DurationMs:  time.Since(t0).Milliseconds(),
-		Message:     writerResult.Message,
+		Message:     msg,
 	})
 
-	b, err := json.Marshal(writerResult)
-	if err != nil {
-		return nil, fmt.Errorf("marshal result: %w", err)
-	}
-	return b, nil
+	return json.Marshal(map[string]any{
+		"reader_analysis": readerText,
+		"writer_output":   writerText,
+		"model":           model,
+		"duration_ms":     dur,
+	})
 }

internal/skills/trainer/handlers_test.go

@@ -6,7 +6,6 @@ import (
 	"encoding/json"
 	"testing"
 
-	iexec "github.com/mathiasbq/supervisor/internal/exec"
 	"github.com/mathiasbq/supervisor/internal/session"
 	"github.com/mathiasbq/supervisor/internal/skills/trainer"
 	"github.com/stretchr/testify/assert"
@@ -31,52 +30,44 @@ func TestTrainerRequiresSessionID(t *testing.T) {
 func TestTrainerCallsReaderThenWriter(t *testing.T) {
 	sessDir := t.TempDir()
 	require.NoError(t, session.Append(sessDir, "sess-1", session.Entry{
-		SessionID: "sess-1", Skill: "tdd", Phase: "red", FinalStatus: "pass",
+		SessionID: "sess-1", Skill: "tdd", Phase: "red", FinalStatus: "ok",
 		Message: "wrote failing test", FilePath: "internal/foo/foo_test.go",
 	}))
 
 	callCount := 0
 	var readerTask, writerTask string
 
-	fakeFn := func(_ context.Context, req iexec.Request) (iexec.Result, error) {
+	fakeFn := func(_ context.Context, _, sys, user string) (string, int64, error) {
 		callCount++
 		if callCount == 1 {
 			// reader call
-			readerTask = req.TaskPrompt
-			return iexec.Result{
-				Status: "pass", Phase: "trainer", Skill: "trainer",
-				RunnerOutput: `[{"type":"sft","moment":"first-pass clean TDD","score":4}]`,
-				Verified: true, ModelUsed: "self", Message: "1 sft candidate found",
-			}, nil
+			readerTask = user
+			return "1 sft candidate found: first-pass clean TDD", 60, nil
 		}
 		// writer call
-		writerTask = req.TaskPrompt
-		return iexec.Result{
-			Status: "pass", Phase: "trainer", Skill: "trainer",
-			FilePath: sessDir + "/training-data/sft/sess-1.jsonl",
-			Verified: true, ModelUsed: "self", Message: "1 sft pair written",
-		}, nil
+		writerTask = user
+		return "written 1 knowledge entry to brain/knowledge/tdd-patterns.md", 70, nil
 	}
 
 	sk := trainer.New(trainer.Config{
 		ReaderPrompt: "reader rules",
 		WriterPrompt: "writer rules",
-		ExecutorFn:   fakeFn,
+		CompleteFunc: fakeFn,
 		SessionsDir:  sessDir,
 		BrainDir:     t.TempDir(),
 	})
 	out, err := sk.Handle(context.Background(), "trainer", json.RawMessage(`{"session_id":"sess-1"}`))
 	require.NoError(t, err)
 
-	assert.Equal(t, 2, callCount, "executor must be called exactly twice: reader then writer")
+	assert.Equal(t, 2, callCount, "complete must be called exactly twice: reader then writer")
 	assert.Contains(t, readerTask, "role: reader")
 	assert.Contains(t, readerTask, "sess-1")
-	assert.Contains(t, readerTask, "wrote failing test") // session history in reader prompt
+	assert.Contains(t, readerTask, "wrote failing test")
 	assert.Contains(t, writerTask, "role: writer")
-	assert.Contains(t, writerTask, "sft candidate") // reader output passed to writer
+	assert.Contains(t, writerTask, "sft candidate")
 
-	var result iexec.Result
+	var result map[string]any
 	require.NoError(t, json.Unmarshal(out, &result))
-	assert.Equal(t, "trainer", result.Phase)
-	assert.Equal(t, "pass", result.Status)
+	assert.Contains(t, result["reader_analysis"], "sft candidate")
+	assert.Contains(t, result["writer_output"], "knowledge entry")
 }

internal/skills/trainer/skill.go

@@ -5,21 +5,20 @@ import (
 	"context"
 	"encoding/json"
 
-	iexec "github.com/mathiasbq/supervisor/internal/exec"
 	"github.com/mathiasbq/supervisor/internal/registry"
 )
 
-// ExecutorFn is the function signature for running a worker subprocess.
-type ExecutorFn func(ctx context.Context, req iexec.Request) (iexec.Result, error)
+// CompleteFunc is the function used to call a local model.
+type CompleteFunc func(ctx context.Context, model, system, user string) (string, int64, error)
 
 // Config holds dependencies for the trainer skill.
 type Config struct {
 	ReaderPrompt string
 	WriterPrompt string
 	DefaultModel string
-	ExecutorFn   ExecutorFn
+	CompleteFunc CompleteFunc
 	SessionsDir  string
-	BrainDir     string // root of brain/ directory; writer writes to BrainDir/training-data/
+	BrainDir     string // root of brain/ directory
 }
 
 // Skill implements the trainer MCP tool.
@@ -40,7 +39,7 @@ func (s *Skill) Tools() []registry.ToolDef {
 	return []registry.ToolDef{
 		{
 			Name:        "trainer",
-			Description: "Extract SFT and DPO training pairs from a session log. Runs a reader→writer chain: reader identifies learning moments, writer formats and writes pairs to brain/training-data/.",
+			Description: "Consult a local model to identify learning moments from a session log and suggest knowledge to preserve in the brain.",
 			InputSchema: schema(
 				[]string{"session_id"},
 				map[string]any{