alloy-runtime-cli 0.2.76__tar.gz → 0.2.78__tar.gz

{alloy_runtime_cli-0.2.76 → alloy_runtime_cli-0.2.78}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: alloy-runtime-cli
-Version: 0.2.76
+Version: 0.2.78
 Summary: Alloy Runtime CLI - Command-line interface for Alloy Runtime server
 Keywords: alloy,cli,python,terminal,workflow
 Classifier: Development Status :: 3 - Alpha
@@ -16,8 +16,8 @@ Requires-Dist: textual==8.2.3
 Requires-Dist: genson==1.3.0
 Requires-Dist: pyperclip==1.11.0
 Requires-Dist: prompt-toolkit==3.0.52
-Requires-Dist: alloy-runtime-sdk==0.2.76
-Requires-Dist: alloy-runtime-types==0.2.76
+Requires-Dist: alloy-runtime-sdk==0.2.78
+Requires-Dist: alloy-runtime-types==0.2.78
 Requires-Dist: trogon==0.6.0
 
 # alloy-runtime-cli
@@ -77,7 +77,7 @@ alloy executions get <execution_id> --watch
 alloy executions get <execution_id> --full-output
 ```
 
-`alloy executions get` uses a dedicated transcript view for human output: one metadata header followed by chronological reasoning, grouped tool-call/result, and assistant blocks. Tool inputs are always complete; tool outputs are compact by default, and `--full-output` expands them. `--watch` repaints a live grouped snapshot while the execution is still running and waits indefinitely unless you pass `--timeout <seconds>`.
+`alloy executions get` uses a dedicated transcript view for human output: one metadata header followed by chronological reasoning, grouped tool-call/result, and assistant blocks. Tool inputs are always complete; tool outputs are compact by default, and `--full-output` expands them. RAG retrieval diagnostics are printed above `rag_search` tool output bodies, with failures highlighted red and degraded retrieval highlighted yellow. `--watch` repaints a live grouped snapshot while the execution is still running and waits indefinitely unless you pass `--timeout <seconds>`.
 
 `alloy ping` is the authenticated connectivity check. It confirms who you are, which host you reached, and which server build answered the request.
 

{alloy_runtime_cli-0.2.76 → alloy_runtime_cli-0.2.78}/README.md

@@ -55,7 +55,7 @@ alloy executions get <execution_id> --watch
 alloy executions get <execution_id> --full-output
 ```
 
-`alloy executions get` uses a dedicated transcript view for human output: one metadata header followed by chronological reasoning, grouped tool-call/result, and assistant blocks. Tool inputs are always complete; tool outputs are compact by default, and `--full-output` expands them. `--watch` repaints a live grouped snapshot while the execution is still running and waits indefinitely unless you pass `--timeout <seconds>`.
+`alloy executions get` uses a dedicated transcript view for human output: one metadata header followed by chronological reasoning, grouped tool-call/result, and assistant blocks. Tool inputs are always complete; tool outputs are compact by default, and `--full-output` expands them. RAG retrieval diagnostics are printed above `rag_search` tool output bodies, with failures highlighted red and degraded retrieval highlighted yellow. `--watch` repaints a live grouped snapshot while the execution is still running and waits indefinitely unless you pass `--timeout <seconds>`.
 
 `alloy ping` is the authenticated connectivity check. It confirms who you are, which host you reached, and which server build answered the request.
 

{alloy_runtime_cli-0.2.76 → alloy_runtime_cli-0.2.78}/alloy_runtime_cli.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: alloy-runtime-cli
-Version: 0.2.76
+Version: 0.2.78
 Summary: Alloy Runtime CLI - Command-line interface for Alloy Runtime server
 Keywords: alloy,cli,python,terminal,workflow
 Classifier: Development Status :: 3 - Alpha
@@ -16,8 +16,8 @@ Requires-Dist: textual==8.2.3
 Requires-Dist: genson==1.3.0
 Requires-Dist: pyperclip==1.11.0
 Requires-Dist: prompt-toolkit==3.0.52
-Requires-Dist: alloy-runtime-sdk==0.2.76
-Requires-Dist: alloy-runtime-types==0.2.76
+Requires-Dist: alloy-runtime-sdk==0.2.78
+Requires-Dist: alloy-runtime-types==0.2.78
 Requires-Dist: trogon==0.6.0
 
 # alloy-runtime-cli
@@ -77,7 +77,7 @@ alloy executions get <execution_id> --watch
 alloy executions get <execution_id> --full-output
 ```
 
-`alloy executions get` uses a dedicated transcript view for human output: one metadata header followed by chronological reasoning, grouped tool-call/result, and assistant blocks. Tool inputs are always complete; tool outputs are compact by default, and `--full-output` expands them. `--watch` repaints a live grouped snapshot while the execution is still running and waits indefinitely unless you pass `--timeout <seconds>`.
+`alloy executions get` uses a dedicated transcript view for human output: one metadata header followed by chronological reasoning, grouped tool-call/result, and assistant blocks. Tool inputs are always complete; tool outputs are compact by default, and `--full-output` expands them. RAG retrieval diagnostics are printed above `rag_search` tool output bodies, with failures highlighted red and degraded retrieval highlighted yellow. `--watch` repaints a live grouped snapshot while the execution is still running and waits indefinitely unless you pass `--timeout <seconds>`.
 
 `alloy ping` is the authenticated connectivity check. It confirms who you are, which host you reached, and which server build answered the request.
 

{alloy_runtime_cli-0.2.76 → alloy_runtime_cli-0.2.78}/alloy_runtime_cli.egg-info/requires.txt

@@ -4,6 +4,6 @@ textual==8.2.3
 genson==1.3.0
 pyperclip==1.11.0
 prompt-toolkit==3.0.52
-alloy-runtime-sdk==0.2.76
-alloy-runtime-types==0.2.76
+alloy-runtime-sdk==0.2.78
+alloy-runtime-types==0.2.78
 trogon==0.6.0

{alloy_runtime_cli-0.2.76 → alloy_runtime_cli-0.2.78}/cli/commands/executions/get/transcript_renderer.py

@@ -270,6 +270,9 @@ def _format_rag_search(
         )
         inline_meta = f"{hits} hits" + (f" · {timing}ms" if timing is not None else "")
         body, _ = _format_generic(persisted_output, full=full, width=width)
+        diagnostic_line = _format_retrieval_diagnostic_line(metadata)
+        if diagnostic_line:
+            body = f"{diagnostic_line}\n{body}" if body else diagnostic_line
         return body, inline_meta
 
     titles = _list_of_strings(payload.get("top_titles"))
@@ -315,6 +318,35 @@ def _format_rag_search(
     return "\n".join(lines), inline_meta
 
 
+def _format_retrieval_diagnostic_line(metadata: dict[str, Any]) -> str | None:
+    raw_diagnostics = metadata.get("retrieval_diagnostics")
+    if not isinstance(raw_diagnostics, dict):
+        return None
+
+    diagnostics = cast(dict[str, Any], raw_diagnostics)
+    status = diagnostics.get("status")
+    if status not in {"failed", "degraded"}:
+        return None
+
+    codes = _retrieval_diagnostic_codes(diagnostics.get("warnings"))
+    label = "retrieval failed" if status == "failed" else "retrieval degraded"
+    suffix = f": {', '.join(codes)}" if codes else ""
+    style = "bold red" if status == "failed" else "yellow"
+    return "  " + _styled(escape(label + suffix), style)
+
+
+def _retrieval_diagnostic_codes(value: Any) -> list[str]:
+    warnings = cast(list[object], value) if isinstance(value, list) else []
+    codes: list[str] = []
+    for warning in warnings:
+        if not isinstance(warning, dict):
+            continue
+        raw_code = cast(dict[str, Any], warning).get("code")
+        if isinstance(raw_code, str) and raw_code:
+            codes.append(raw_code)
+    return codes
+
+
 def _format_generic(
     payload: dict[str, Any] | list[Any] | str, *, full: bool, width: int
 ) -> tuple[str, str | None]:

{alloy_runtime_cli-0.2.76 → alloy_runtime_cli-0.2.78}/cli/commands/knowledge/collections/status/presenter.py

@@ -34,6 +34,13 @@ def present_collection_statuses(
             compact_line=1,
             compact_style="yellow",
         ),
+        ColumnConfig(
+            source_field="staged_documents",
+            header_label="Staged",
+            align="right",
+            compact_line=1,
+            compact_style="yellow",
+        ),
         ColumnConfig(
             source_field="processing_documents",
             header_label="Processing",

{alloy_runtime_cli-0.2.76 → alloy_runtime_cli-0.2.78}/cli/commands/knowledge/documents/ingest/command.py

@@ -1,7 +1,7 @@
 """Knowledge document ingest command implementation."""
 
 from pathlib import Path
-from typing import Annotated, Optional
+from typing import Annotated, Optional, cast, Any
 
 import typer
 
@@ -14,7 +14,10 @@ from cli.commands.knowledge.documents.ingest.presenter import present_ingest_suc
 from cli.infrastructure.command import async_command, authenticated_client
 from cli.infrastructure.file_content import FileContentError, resolve_content
 from cli.infrastructure.kv_parser import parse_kv_pairs
-from alloy_runtime_types.dtos.knowledge import IngestDocumentRequest
+from alloy_runtime_types.dtos.knowledge import (
+    IngestDocumentItem,
+    IngestDocumentsRequest,
+)
 
 
 def knowledge_documents_ingest_command(
@@ -89,11 +92,6 @@ def knowledge_documents_ingest_command(
         "--external-id",
         help="Caller-provided ID for idempotency and correlation (unique per org)",
     ),
-    dry_run: bool = typer.Option(
-        False,
-        "--dry-run",
-        help="Only preprocess - don't persist. Useful for testing preprocessing.",
-    ),
     skip_question_gen: bool = typer.Option(
         False,
         "--skip-question-gen",
@@ -118,7 +116,6 @@ def knowledge_documents_ingest_command(
         cat doc.txt | alloy knowledge documents ingest -C <uuid> -t "Doc" -c @- -Q <agent> -T <template>
         alloy knowledge documents ingest -C <uuid> -t "Doc" -c @notes.md -Q <agent> -T <template> -m author=John
         alloy knowledge documents ingest -C <uuid> -t "Meeting" -f meeting.vtt -D transcript -Q <agent> -T <template>
-        alloy knowledge documents ingest -C <uuid> -t "Test" -f doc.md --dry-run -Q <agent> -T <template>
     """
     require_one_of(("--file", file), ("--content", content))
 
@@ -134,7 +131,6 @@ def knowledge_documents_ingest_command(
         tag=tag,
         document_type=document_type,
         external_id=external_id,
-        dry_run=dry_run,
         skip_question_gen=skip_question_gen,
     )
 
@@ -151,7 +147,6 @@ def _execute_ingest_from_flags(
     tag: Optional[list[str]],
     document_type: str,
     external_id: Optional[str],
-    dry_run: bool,
     skip_question_gen: bool,
 ) -> None:
     """Build and execute document ingestion from command flags."""
@@ -194,8 +189,7 @@ def _execute_ingest_from_flags(
     # Parse metadata (httpie style)
     parsed_metadata = parse_kv_pairs(metadata) if metadata else None
 
-    request = IngestDocumentRequest(
-        collection_id=collection,
+    item = IngestDocumentItem(
         title=title,
         content=document_content,
         question_gen_agent_id=question_agent,
@@ -205,17 +199,17 @@ def _execute_ingest_from_flags(
         metadata=parsed_metadata,
         tags=tag,
         external_id=external_id,
-        dry_run=dry_run,
         skip_question_generation=skip_question_gen,
     )
+    request = IngestDocumentsRequest(collection_id=collection, documents=[item])
 
     _execute_ingest(request)
 
 
 @async_command
-async def _execute_ingest(request: IngestDocumentRequest) -> None:
+async def _execute_ingest(request: IngestDocumentsRequest) -> None:
     """Execute the document ingestion with the given request."""
     async with authenticated_client() as (_config, client):
-        response = await client.ingest_document(request)
+        response = await cast(Any, client).ingest_document(request)
 
     present_ingest_success(response)

alloy_runtime_cli-0.2.78/cli/commands/knowledge/documents/ingest/presenter.py

@@ -0,0 +1,29 @@
+"""Presenter for document ingest output."""
+
+from cli.infrastructure.output import OutputService
+from alloy_runtime_types.dtos.knowledge import IngestDocumentsResponse
+
+
+def present_ingest_success(response: IngestDocumentsResponse) -> None:
+    """Present successful document ingestion (queued for processing).
+
+    Args:
+        response: Document ingestion response from server
+    """
+    output = OutputService.get()
+    extra_fields: dict[str, object] = {
+        "Submitted": response.submitted_count,
+        "Staged": response.staged_count,
+        "Duplicate external id": response.duplicate_external_id_count,
+        "Duplicate content": response.duplicate_content_count,
+        "Duplicate within batch": response.duplicate_within_batch_count,
+        "Next Step": f"poll GET /knowledge/documents/ingest/operations/{response.operation_id}",
+    }
+    output.result(
+        action="ingest",
+        resource="knowledge_documents",
+        status=response.status,
+        identifier=str(response.operation_id),
+        name=None,
+        extra_fields=extra_fields or None,
+    )

{alloy_runtime_cli-0.2.76 → alloy_runtime_cli-0.2.78}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "alloy-runtime-cli"
-version = "0.2.76"
+version = "0.2.78"
 description = "Alloy Runtime CLI - Command-line interface for Alloy Runtime server"
 readme = "README.md"
 requires-python = ">=3.13"
@@ -23,8 +23,8 @@ dependencies = [
   "genson==1.3.0",
   "pyperclip==1.11.0",
   "prompt-toolkit==3.0.52",
-  "alloy-runtime-sdk==0.2.76",
-  "alloy-runtime-types==0.2.76",
+  "alloy-runtime-sdk==0.2.78",
+  "alloy-runtime-types==0.2.78",
   "trogon==0.6.0",
 ]
 

alloy_runtime_cli-0.2.76/cli/commands/knowledge/documents/ingest/presenter.py

@@ -1,28 +0,0 @@
-"""Presenter for document ingest output."""
-
-from cli.infrastructure.output import OutputService
-from alloy_runtime_types.dtos.knowledge import IngestDocumentResponse
-
-
-def present_ingest_success(response: IngestDocumentResponse) -> None:
-    """Present successful document ingestion (queued for processing).
-
-    Args:
-        response: Document ingestion response from server
-    """
-    output = OutputService.get()
-    extra_fields: dict[str, object] = {}
-    if response.document_id is not None:
-        extra_fields["Next Step"] = (
-            f"alloy knowledge documents get {response.document_id}"
-        )
-    output.result(
-        action="ingest",
-        resource="knowledge_document",
-        status=response.status,
-        identifier=(
-            str(response.document_id) if response.document_id is not None else None
-        ),
-        name=response.title,
-        extra_fields=extra_fields or None,
-    )