fix(store): unique event IDs on lease/expiry, honest duplicate-ingest signal (S5, S6)

S5: Store.Lease and Store.ExpireLeases both set Event.ID to the task id, so
every TaskLeased/TaskReleased event for a given task collided on ID across
every lease of that task — unsound for ApplyAdvisory or any future
ID-based lookup. Both now call domain.NewID().

S6: Append's TaskCreated dedup path returned nil (success) without
appending anything. main.go's handler then did
`s.Events(0)[len(s.Events(0))-1]` and returned that — an unrelated event —
with 201 Created, and every other Append caller (Gitea poll/webhook, JSONL
ingest) had no way to distinguish "duplicate, as expected" from "genuinely
appended".

Add domain.ErrDuplicate, returned instead of nil on a duplicate
(source, external_id). Add Store.TaskBySource to resolve the
already-ingested task by that same dedup key. Update every caller:
  - main.go's POST /v1/tasks now returns 200 with the existing task on
    ErrDuplicate instead of fabricating a 201 with the wrong event.
  - provider.Gitea.Poll/IngestWebhook and provider.JSONL.Ingest treat
    ErrDuplicate as expected (already-seen issue/line), not a failure —
    without this, Gitea polling would have errored out of its loop on the
    first already-ingested issue in every batch, since Poll previously
    relied on the old nil-on-dup behavior to keep scanning.

TestLeaseAndExpireEventIDsAreUnique and TestTaskBySourceResolvesDuplicate
cover the store-level fixes; TestAppendReplayAndDeduplicate updated for the
new error signal.

AUDIT.md S5, S6.

cmd/orchestra/main.go

@@ -3,6 +3,7 @@ package main
 import (
 	"context"
 	"encoding/json"
+	"errors"
 	"io"
 	"log"
 	"net"
@@ -168,6 +169,18 @@ func main() {
 		b, _ := json.Marshal(p)
 		e := domain.Event{ID: id(), Type: "TaskCreated", TaskID: id(), Version: 1, Payload: b, Surface: string(surface(r))}
 		if err := s.Append(e); err != nil {
+			if errors.Is(err, domain.ErrDuplicate) {
+				// (source, external_id) was already ingested. Nothing was
+				// appended; report the existing task idempotently rather
+				// than fabricating a 201 with an unrelated event (S6).
+				source, _ := p["source"].(string)
+				externalID, _ := p["external_id"].(string)
+				if t, ok := s.TaskBySource(source, externalID); ok {
+					w.WriteHeader(200)
+					json.NewEncoder(w).Encode(t)
+					return
+				}
+			}
 			http.Error(w, err.Error(), 400)
 			return
 		}
@@ -176,7 +189,6 @@ func main() {
 				log.Printf("route task: %v", err)
 			}
 		}
-		e = s.Events(0)[len(s.Events(0))-1]
 		w.WriteHeader(201)
 		json.NewEncoder(w).Encode(e)
 	})

internal/domain/domain.go

@@ -17,6 +17,11 @@ var ErrConflict = errors.New("task version conflict")
 var ErrNotFound = errors.New("task not found")
 var ErrInvalid = errors.New("invalid event")
 
+// ErrDuplicate is returned by Store.Append for a TaskCreated event whose
+// (source, external_id) pair was already ingested. The caller already has a
+// task for this content; nothing was appended.
+var ErrDuplicate = errors.New("duplicate task ingestion")
+
 // CurrentEventSchema is 2: schema 2 requires every event to declare its
 // authorizing Surface (see ValidateEvent), enforced at the store append
 // boundary. Schema 1 events already on disk replay unchanged — tolerant

internal/provider/provider.go

@@ -137,10 +137,13 @@ func (j JSONL) Ingest(r io.Reader, sink Sink) (int, error) {
 			return count, fmt.Errorf("line %d: %w", line, err)
 		}
 		b, _ := json.Marshal(p)
-		if err := sink.Append(domain.Event{ID: domain.NewID(), Type: "TaskCreated", TaskID: domain.NewID(), Version: 1, Payload: b, Surface: string(authz.System)}); err != nil {
+		err := sink.Append(domain.Event{ID: domain.NewID(), Type: "TaskCreated", TaskID: domain.NewID(), Version: 1, Payload: b, Surface: string(authz.System)})
+		if err != nil && !errors.Is(err, domain.ErrDuplicate) {
 			return count, fmt.Errorf("line %d: %w", line, err)
 		}
-		count++
+		if err == nil {
+			count++
+		}
 	}
 	return count, sc.Err()
 }
@@ -321,7 +324,10 @@ func (g Gitea) IngestWebhook(body []byte, signature string, sink Sink) error {
 			project = h.Repository.FullName
 		}
 	}
-	return sink.Append(g.event(h.Issue, g.sourceName(), project))
+	if err := sink.Append(g.event(h.Issue, g.sourceName(), project)); err != nil && !errors.Is(err, domain.ErrDuplicate) {
+		return err
+	}
+	return nil
 }
 func validSignature(body []byte, got, secret string) bool {
 	if secret == "" || got == "" {
@@ -379,7 +385,7 @@ func (g Gitea) Poll(ctx context.Context, sink Sink) (int, error) {
 		project = g.Repo
 	}
 	for _, i := range issues {
-		if err := sink.Append(g.event(i, g.sourceName(), project)); err != nil {
+		if err := sink.Append(g.event(i, g.sourceName(), project)); err != nil && !errors.Is(err, domain.ErrDuplicate) {
 			return 0, err
 		}
 	}

internal/store/store.go

@@ -189,7 +189,7 @@ func (s *Store) Append(e domain.Event) error {
 			return err
 		}
 		if id := s.external[p["source"].(string)+"\x00"+p["external_id"].(string)]; id != "" {
-			return nil
+			return domain.ErrDuplicate
 		}
 	}
 	t, taskExists := s.tasks[e.TaskID]
@@ -326,6 +326,19 @@ func (s *Store) Task(id string) (domain.Task, bool) {
 	return t, ok
 }
 
+// TaskBySource resolves the task ingested for a given (source, external_id)
+// pair — the dedup key Append.ErrDuplicate rejects re-ingestion against.
+func (s *Store) TaskBySource(source, externalID string) (domain.Task, bool) {
+	s.mu.Lock()
+	defer s.mu.Unlock()
+	id := s.external[source+"\x00"+externalID]
+	if id == "" {
+		return domain.Task{}, false
+	}
+	t, ok := s.tasks[id]
+	return t, ok
+}
+
 func (s *Store) Lease(id, harness string, ttl time.Duration) (domain.Event, error) {
 	if ttl <= 0 {
 		return domain.Event{}, fmt.Errorf("%w: ttl must be positive", domain.ErrInvalid)
@@ -338,7 +351,7 @@ func (s *Store) Lease(id, harness string, ttl time.Duration) (domain.Event, erro
 		return domain.Event{}, domain.ErrConflict
 	}
 	p, _ := json.Marshal(map[string]any{"harness_id": harness, "ttl": ttl.Seconds(), "until_ns": time.Now().Add(ttl).UnixNano(), "expected_version": t.Version})
-	e := domain.Event{ID: id, Type: "TaskLeased", TaskID: id, Version: t.Version + 1, Payload: p, Surface: string(authz.System)}
+	e := domain.Event{ID: domain.NewID(), Type: "TaskLeased", TaskID: id, Version: t.Version + 1, Payload: p, Surface: string(authz.System)}
 	return e, s.Append(e)
 }
 
@@ -347,7 +360,7 @@ func (s *Store) ExpireLeases(now time.Time) ([]domain.Event, error) {
 	for _, t := range s.Tasks() {
 		if t.State == domain.StateLeased && t.Lease != nil && !t.Lease.Until.After(now) {
 			p, _ := json.Marshal(map[string]any{"reason": "lease_expired", "harness_id": t.Lease.HarnessID})
-			e := domain.Event{ID: t.ID, Type: "TaskReleased", TaskID: t.ID, Version: t.Version + 1, Payload: p, Surface: string(authz.System)}
+			e := domain.Event{ID: domain.NewID(), Type: "TaskReleased", TaskID: t.ID, Version: t.Version + 1, Payload: p, Surface: string(authz.System)}
 			if err := s.Append(e); err != nil {
 				return out, err
 			}

internal/store/store_test.go

@@ -2,9 +2,11 @@ package store
 
 import (
 	"encoding/json"
+	"errors"
 	"os"
 	"path/filepath"
 	"testing"
+	"time"
 
 	"orchestra/internal/authz"
 	"orchestra/internal/domain"
@@ -24,8 +26,8 @@ func TestAppendReplayAndDeduplicate(t *testing.T) {
 	if err := s.Append(created("e1")); err != nil {
 		t.Fatal(err)
 	}
-	if err := s.Append(created("e2")); err != nil {
+	if err := s.Append(created("e2")); !errors.Is(err, domain.ErrDuplicate) {
-		t.Fatal(err)
+		t.Fatalf("expected ErrDuplicate, got %v", err)
 	}
 	if got := len(s.Events(0)); got != 1 {
 		t.Fatalf("duplicate ingest appended %d events", got)
@@ -53,6 +55,63 @@ func TestAppendReplayAndDeduplicate(t *testing.T) {
 	}
 }
 
+// TestLeaseAndExpireEventIDsAreUnique guards S5: Event.ID was set to the
+// task id in both Lease and ExpireLeases, so every lease of the same task
+// produced a TaskLeased/TaskReleased event with a colliding ID — unsound
+// for ApplyAdvisory or any future ID-based lookup.
+func TestLeaseAndExpireEventIDsAreUnique(t *testing.T) {
+	s, err := Open(t.TempDir())
+	if err != nil {
+		t.Fatal(err)
+	}
+	if err := s.Append(created("e1")); err != nil {
+		t.Fatal(err)
+	}
+	task := s.Tasks()[0]
+	leaseEvt, err := s.Lease(task.ID, "h1", time.Millisecond)
+	if err != nil {
+		t.Fatal(err)
+	}
+	if leaseEvt.ID == task.ID || leaseEvt.ID == "" {
+		t.Fatalf("lease event ID %q collides with task ID %q", leaseEvt.ID, task.ID)
+	}
+	time.Sleep(2 * time.Millisecond)
+	expired, err := s.ExpireLeases(time.Now())
+	if err != nil {
+		t.Fatal(err)
+	}
+	if len(expired) != 1 {
+		t.Fatalf("expected 1 expiry, got %d", len(expired))
+	}
+	if expired[0].ID == task.ID || expired[0].ID == leaseEvt.ID || expired[0].ID == "" {
+		t.Fatalf("expiry event ID %q collides", expired[0].ID)
+	}
+}
+
+// TestTaskBySourceResolvesDuplicate covers the S6 fix: a caller that gets
+// ErrDuplicate from Append must be able to look up the already-ingested
+// task by its dedup key instead of guessing at "the last event in the log".
+func TestTaskBySourceResolvesDuplicate(t *testing.T) {
+	s, err := Open(t.TempDir())
+	if err != nil {
+		t.Fatal(err)
+	}
+	if err := s.Append(created("e1")); err != nil {
+		t.Fatal(err)
+	}
+	want := s.Tasks()[0]
+	if err := s.Append(created("e2")); !errors.Is(err, domain.ErrDuplicate) {
+		t.Fatalf("expected ErrDuplicate, got %v", err)
+	}
+	got, ok := s.TaskBySource("jsonl", "42")
+	if !ok || got.ID != want.ID {
+		t.Fatalf("TaskBySource = %+v, ok=%v, want %+v", got, ok, want)
+	}
+	if _, ok := s.TaskBySource("jsonl", "does-not-exist"); ok {
+		t.Fatal("expected not found")
+	}
+}
+
 func TestArtifactIsContentAddressed(t *testing.T) {
 	s, err := Open(t.TempDir())
 	if err != nil {