You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Commit 0bc58d1
Browse filesBrowse the repository at this point in the historyBrowse files
-**ParseX parsing**: Add `bl parsex parse submit/result` for document and audio/video parsing, result queries, and Markdown output.
14
+
-**ParseX field extraction**: Add `bl parsex extract submit/result` to extract fields with a JSON Schema from documents or existing parsing results, including source citations.
15
+
-**Configuration source diagnostics**: Show API key and Base URL sources in `bl config show`, and warn when requests use values from different configuration sources.
16
+
17
+
### Fixed
18
+
19
+
-**Workspace API key login**: Fix `bl auth login` rejecting valid workspace API keys on MaaS endpoints.
20
+
-**Login validation errors**: Preserve service and network errors during API key validation instead of reporting every failure as an invalid key.
21
+
22
+
## [2.6.2] - 2026-10-28
23
+
24
+
### Changed
25
+
26
+
-**Default telemetry output (BREAKING)**: Business results from `bl alert`, `bl monitor`, and `bl log` now default to JSON, even when the Profile specifies text. Scripts relying on text must pass `--output text` or set `DASHSCOPE_OUTPUT=text`.
27
+
28
+
### Fixed
29
+
30
+
-**Workspace identity**: Resolve the selected workspace's Console identity automatically and share it across preflight, business requests, asynchronous polling, and follow-up requests, fixing ineffective `--workspace-id` filtering and incorrect cross-workspace results without changing saved configuration.
31
+
-**Audit log parsing**: Normalize timestamps, models, request IDs, workspace IDs, and usage; fix missing fields, NaN timestamps, and malformed count displays, and reject audit results from a mismatched workspace.
32
+
-**Log counts**: Use each corresponding list API's `totalCount` for `log audit count` and `log inference count`, fixing zero counts for non-empty lists without paginating or fetching full log content.
33
+
-**Inference count filters and completion**: Preserve multiple-model and multiple-API-key filters by deduplicating and querying single-value combinations. Support asynchronous polling and emit totals only after every query succeeds; reject empty filters, invalid counts, and aggregate overflow.
34
+
-**Dry-run previews**: Show the workspace identity resolution plan and actual count request combinations without sending business requests.
35
+
36
+
## [2.6.1] - 2026-10-28
37
+
38
+
### Added
39
+
40
+
-**Asynchronous Batch inference**: Add `bl batch convert/create/list/get/cancel/download` for CSV-to-JSONL conversion, Chat/Embedding task creation, progress tracking, cancellation, and result/error downloads with optional CSV export.
41
+
-**Synchronous Batch Chat**: Add `bl text chat --batch` for one non-streaming request, separate from asynchronous file-based Batch tasks, with request timeouts of 60–3600 seconds.
42
+
-**Model recommendation options**: Add `--top`, `--model`, `--no-agent`, `--trace`, `--think`, and `--no-thinking` to `bl advisor recommend`.
43
+
44
+
### Changed
45
+
46
+
-**Model recommendations**: Upgrade intent analysis, routing, recall, scoring, and ranking; support Agent clarification in interactive terminals and multi-step recommendations while preserving the existing recommendation JSON envelope.
47
+
48
+
### Removed
49
+
50
+
-**Advisor library interfaces (BREAKING)**: Remove legacy recommendation exports, including `getModels`, `recallCandidates`, `recallSemantic`, and `rankModels`. Library consumers must adapt to the new recommendation engine; legacy types are exported as `LegacyIntentProfile`, `LegacyModelProfile`, and `LegacyRecommendResult`.
51
+
52
+
### Fixed
53
+
54
+
-**Temporary file uploads**: Select the upload service from the active model endpoint, fixing Singapore and Prime video uploads incorrectly using Beijing. Official temporary uploads support only Beijing and Singapore; unsupported regions and Token Plan fail explicitly, while explicit custom upload proxies are no longer rejected based on key type.
55
+
-**Node 26 compatibility**: Fix fetch/dispatcher incompatibility and proxy initialization across model requests, console requests, uploads, and downloads.
56
+
-**API error passthrough**: Recognize nested business errors and preserve non-JSON error bodies and request IDs instead of treating error responses as successful results.
57
+
58
+
## [2.6.0] - 2026-10-27
59
+
60
+
### Added
61
+
62
+
-**Knowledge initialization**: Add `knowledge init` to upload a sample, create a knowledge base and retrieval service, and verify retrieval, with recovery and resource reuse.
63
+
-**Directory sync and file import**: Add `knowledge doc sync` and `knowledge doc import` for local directories and existing files.
64
+
-**Multimedia knowledge bases**: Support audio/video knowledge bases, parser configuration, streaming uploads, image-only search, and online search filters.
65
+
-**Agent capability discovery**: Add `--introspect` and the `bailian-knowledge` skill, with corresponding capabilities in Knowledge Studio CLI.
66
+
67
+
### Changed
68
+
69
+
-**Chat output**: Separate answers, phases, tools, sources, and usage in structured output; quiet mode emits only the final answer.
70
+
-**Knowledge-base creation**: Add billing notices, resource records, and cleanup guidance.
71
+
72
+
### Fixed
73
+
74
+
- Reject blank required inputs in knowledge commands.
75
+
- Report completed, unconfirmed, and unexecuted batches when chunk deletion fails.
76
+
77
+
## [2.5.1] - 2026-10-23
78
+
79
+
### Added
80
+
81
+
-**Console login**: Support `--no-browser` for manually opening the login URL and `--token-stdin` for importing and validating console credentials from standard input.
82
+
-**Agent attribution**: Support declaring the calling Agent, Skill name and version; undeclared callers remain unknown.
83
+
84
+
### Fixed
85
+
86
+
-**Login completion**: Handle fast browser callbacks reliably and report a timeout when no credentials are received.
87
+
-**Credential refresh**: Correct region selection and validate the retried request before saving refreshed credentials.
Copy file name to clipboardExpand all lines: README.md
+48Lines changed: 48 additions & 0 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -28,6 +28,7 @@ _Built for AI Agents. Every command works as a structured tool call._
28
28
-**App orchestration** — Call Managed Agents, agents, and workflows published on Aliyun Model Studio, wired to knowledge bases, memory, web search, and MCP tools
29
29
-**Native MCP setup** — Register Bailian MCP servers in Codex, Claude Code, Cursor, Qoder, Qoder Work, QwenWork (千问办公), Qwen Code, Gemini CLI, or Pi
30
30
-**Training & deployment** — Validate and upload datasets, fine-tune models, deploy dedicated models as endpoints
31
+
-**Batch inference** — Prepare CSV/JSONL, create and follow asynchronous Chat/Embedding tasks, download original results/errors with optional CSV, or send one synchronous Batch Chat request
31
32
-**Account operations** — Login, UI-based configuration, model marketplace, usage and quota, rate-limit increases, team seat management
32
33
-**Plan onboarding** — Connect subscription plans such as Token Plan to the CLI and common coding agents in one step
33
34
@@ -126,6 +127,53 @@ Once installed, just describe your task to your AI Agent — no need to assemble
126
127
127
128
> More examples and scenarios: [Aliyun Model Studio CLI Site](https://bailian.console.aliyun.com/cli?spm=a2ty07.github)
128
129
130
+
### Try a knowledge base
131
+
132
+
With a valid API key and workspace, preview initialization first:
Knowledge bases accrue running-time charges from creation, even without queries. The one-time Standard Edition 720-hour allowance is shared across knowledge bases and expires 30 days after activation for new users. Model calls cost extra; your remaining allowance is unknown. Review the plan and [billing rules](https://help.aliyun.com/zh/model-studio/billing-for-knowledge-base), then add `--yes` to the same command **after confirming the operation**, removing `--dry-run` to execute.
139
+
140
+
Keep the returned IDs and recovery record. Closing the CLI or a later failure does not stop billing; follow the reported cleanup actions to delete unused knowledge bases. See the [Knowledge Studio quick start](packages/kscli/README.md#quick-start) for retrieval and cleanup. The `bailian-knowledge` Skill is available through `bl skill init`.
141
+
142
+
### RAG audio and video
143
+
144
+
`bl knowledge doc import --index-id <id> --doc-id <fileId>` imports an existing file into a knowledge base; `doc list --details` shows per-file processing settings. Media support includes parser options, streaming uploads, and media references in search/chunk output. Chat JSON separates the final answer from phases, tools, docs, usage, and raw events; quiet mode emits only the final answer. See the [command guide](docs/knowledge/knowledge-cli-guide.md).
145
+
146
+
## ParseX
147
+
148
+
Parse documents or audio/video and extract structured fields through the workspace-scoped ParseX API:
Use an ordinary API key and a workspace with ParseX access. Submission returns `data.biz_id` immediately; `result --follow` waits for completion. `schema.json` contains a JSON Schema object, for example `{"type":"object","properties":{"title":{"type":"string"}}}`. Inputs are accessible HTTP(S) URLs; extraction also accepts a retained document parsing task. See the [ParseX command guide](docs/parsex.md) for processing options, result slices, OSS credentials, dry-run and polling behavior.
bl text chat --batch --model qwen-plus --message "Hello" --timeout 3600
169
+
```
170
+
171
+
`--help` includes CSV, JSONL and result data examples. `convert` is local and needs no key; inference/task commands use a model API key. File inference uses asynchronous tasks via `create → get → download`: `create` uploads/submits and returns a task ID without waiting for completion. Query with `get` and download results after completion. Conversion/download files are atomically published per file and never silently overwritten. Use `--force` to recover partial downloads. JSONL is the authoritative result, while CSV is a lossy display; import IDs as text to preserve leading zeros. Ctrl-C stops local polling, not the remote task.
172
+
173
+
`text chat --batch` sends one synchronous request and waits for its response; it does not create an asynchronous task ID. Use `--message` or `--messages-file`, which contains one request's conversation context, not independent file requests. The total request timeout defaults to 300 seconds (`--timeout`: 60–3600 seconds). Timeout/Ctrl-C/disconnection do not guarantee server cancellation.
174
+
175
+
Synchronous Batch Chat is non-streaming; explicit realtime/international/Token Plan endpoints are rejected. An explicit realtime profile requires `--base-url https://batch.dashscope.aliyuncs.com`. Batch availability and limits depend on the service/model. Dry-run avoids business requests and output/directory writes; framework update checks/telemetry follow their settings.
0 commit comments