feat: unified fleet CLI, Main Chat preservation policy, sidechat routing, and file transfers

- Added CHAT_POLICY.md and README.md banner enforcing sidechat-first and file-transfer-first rules.
- Added strict Main Chat block to super dm send and super dm wo with --allow-main-chat override.
- Implemented file transfer staging and metadata registry in super dm send-file and super dm files (with clean subcommand).
- Added full job lifecycle management (show, create, enable, disable, delete, run --follow) to super-cli.py and box-ctl.py.
- Audited all jobs in jobs/*.json and redirected automated dispatches away from Main Chat.
- Hardened chromebox-watchdog.sh with systemd user session environment exports and stale singleton cleanup.
- Added compose_check command choice to muse-chat-api.py.

bin/response-harvester.py

@@ -62,6 +62,12 @@ try:
 except ImportError:
     HAS_REGISTRY = False
 
+try:
+    import pipeline_engine
+    HAS_PIPELINE = True
+except ImportError:
+    HAS_PIPELINE = False
+
 VALID_AGENTS = ["muse", "pip", "646", "opm"]
 DEFAULT_PORTS = {"muse": 9410, "pip": 9420, "646": 9430, "opm": 9440}
 
@@ -384,8 +390,8 @@ def harvest_agent_thread(cdp, agent, thread_info, watermarks, followups, dry_run
                 }
                 if not dry_run:
                     append_jsonl(JOB_LOG, job_record)
-                    # Trigger chain_next if configured
-                    trigger_chain_next(job_id, result_text)
+                    # Trigger pipeline chaining or next job if configured
+                    trigger_chain_next(job_id, result_text, success=not is_fail)
 
             # Check and clear pending follow-ups
             clear_matching_followups(followups, agent, thread_id, mid, text, dry_run)
@@ -393,9 +399,8 @@ def harvest_agent_thread(cdp, agent, thread_info, watermarks, followups, dry_run
     return new_messages, new_wm, job_results
 
 
-def trigger_chain_next(job_id, result_text):
-    """If the completed job has a chain_next property, dispatch it with context."""
-    # Job ID format: <name>-<timestamp>-<uuid>
+def trigger_chain_next(job_id, result_text, success=True):
+    """If the completed job has on_success, on_failure, or chain_next, dispatch downstream."""
     parts = job_id.split("-")
     if len(parts) < 3:
         return
@@ -404,20 +409,55 @@ def trigger_chain_next(job_id, result_text):
     if not job_file.exists():
         return
 
+    # Update pipeline ledger if this job belongs to an active pipeline
+    pipeline_run_id = None
+    step_n = 1
+    if HAS_PIPELINE:
+        run_entry, step_entry = pipeline_engine.record_step_result(job_id, success, result_text)
+        if run_entry:
+            pipeline_run_id = run_entry.get("run_id")
+            step_n = step_entry.get("step_n", 1) + 1
+
     try:
         with open(job_file, "r", encoding="utf-8") as f:
             cfg = json.load(f)
-        chain_next = cfg.get("chain_next")
-        if chain_next and (JOBS_DIR / f"{chain_next}.json").exists():
+
+        next_job = None
+        if success:
+            next_job = cfg.get("on_success") or cfg.get("chain_next")
+        else:
+            next_job = cfg.get("on_failure")
+
+        if next_job and (JOBS_DIR / f"{next_job}.json").exists():
+            # Inter-step settle delay to prevent browser race conditions
+            step_delay = int(cfg.get("step_delay", 5))
+            if step_delay > 0:
+                time.sleep(step_delay)
+
             env = os.environ.copy()
             env["CHAIN_PREV_JOB_ID"] = job_id
             env["CHAIN_PREV_RESULT"] = result_text[:1000]
+            if pipeline_run_id:
+                env["CHAIN_PIPELINE_RUN_ID"] = pipeline_run_id
+                env["CHAIN_STEP_N"] = str(step_n)
+
+            cmd = [sys.executable, str(DISPATCH_PY), next_job]
+            if pipeline_run_id:
+                cmd.extend(["--pipeline-run", pipeline_run_id, "--step-n", str(step_n)])
+
             subprocess.Popen(
-                [sys.executable, str(DISPATCH_PY), chain_next],
+                cmd,
                 env=env,
                 stdout=subprocess.DEVNULL,
                 stderr=subprocess.DEVNULL,
             )
+        else:
+            # End of chain for this pipeline run
+            if pipeline_run_id and HAS_PIPELINE:
+                if success:
+                    pipeline_engine.complete_pipeline(pipeline_run_id)
+                else:
+                    pipeline_engine.fail_pipeline(pipeline_run_id, "step_failed_without_fallback")
     except Exception:
         pass