Baidu Qianfan MCP integration for AI agents.
Connect AI agents to Baidu Qianfan through 25 structured actions, including create ai search completion, create air image generation, cancel batch, and reran…
- ACTION CONTRACTS
- 25
- AUTHENTICATION
- API key
- SOURCE STATUS
- Provider-backed
CATALOG SOURCE REVIEWED AUGUST 23, 2026 / ACTION NAMES AND SCHEMAS DERIVED FROM WORKING MACHINES PROVIDER SOURCE
CAPABILITY PROFILE
What the Baidu Qianfan integration exposes
The Baidu Qianfan Agent App exposes 25 provider-backed actions for ai work. Its current contract lets an authorized agent create a non-streaming ai search chat completion with baidu qianfan, generate images with the baidu qianfan musesteamer air image endpoint, and cancel a baidu qianfan batch prediction job by its identifier. These operations are called through Working Machines as typed capabilities rather than through browser navigation or copied UI steps.
A connection uses API key. Before execution, the agent can inspect the selected action, its required fields, declared scopes, and expected output contract. Provider credentials remain inside the Working Machines runtime; the calling agent receives the capability and its structured result, not the underlying secret.
For reliable operation, start with the narrowest action that satisfies the task, resolve stable provider identifiers before changing state, and validate the returned object or status after execution. Availability still depends on the connected Baidu Qianfan account, granted provider permissions, workspace policy, region, plan, and upstream API behavior.
VERIFIED ACTION SAMPLE
Real Baidu Qianfan capabilities.
Showing 12 of 25 actions. Risk labels are conservative signals based on operation names, not substitutes for provider documentation or runtime policy.
create_ai_search_completionREADCreate a non-streaming AI search chat completion with Baidu Qianfan.
INPUTS: model / messages / stream
create_air_image_generationWRITEGenerate images with the Baidu Qianfan MuseSteamer Air image endpoint.
INPUTS: model / prompt / n / seed / size
cancel_batchHIGH IMPACTCancel a Baidu Qianfan batch prediction job by its identifier.
INPUTS: batch_id
rerankREVIEWScore and rank candidate documents against a query with a Baidu Qianfan rerank model.
INPUTS: model / query / documents / top_n / return_documents
cancel_video_generation_taskHIGH IMPACTCancel a Baidu Qianfan video generation task by its task identifier.
INPUTS: task_id
create_batchWRITECreate a Baidu Qianfan batch prediction job from an uploaded input file.
INPUTS: input_file_id / endpoint / completion_window / metadata / replace
create_chat_completionWRITECreate a non-streaming OpenAI-compatible chat completion with Baidu Qianfan.
INPUTS: model / messages / max_tokens / temperature / top_p
create_completionWRITECreate a non-streaming fill-in-the-middle completion with Baidu Qianfan.
INPUTS: model / prompt / suffix / max_tokens / temperature
create_embeddingsWRITEGenerate embedding vectors for one or more input strings with Baidu Qianfan.
INPUTS: model / input / encoding_format / dimensions / user
create_image_generationWRITEGenerate images with the Baidu Qianfan general image generation endpoint.
INPUTS: model / prompt / n / seed / size
create_responseWRITECreate a non-streaming stored response with the Baidu Qianfan responses API.
INPUTS: model / input / instructions / stream
create_video_generation_taskWRITECreate a Baidu Qianfan video generation task.
INPUTS: model / content / duration / prompt_extend / watermark
INPUT CONTRACTS
Know what the action needs before it runs.
create_ai_search_completion
modelOPTIONAL- The AI search model identifier to use.
messagesOPTIONAL- The ordered conversation messages.
streamOPTIONAL- Whether to request a streaming response. This connector only accepts false or omission.
create_air_image_generation
modelOPTIONAL- The image model identifier to use.
promptOPTIONAL- The prompt used to generate the image.
nOPTIONAL- The number of images to generate.
seedOPTIONAL- The random seed used for generation.
sizeOPTIONAL- The output image size, for example 1024x1024.
cancel_batch
batch_idREQUIRED- The batch identifier.
rerank
modelOPTIONAL- The rerank model identifier.
queryOPTIONAL- The search query used to score the candidate documents.
documentsOPTIONAL- The ordered candidate documents to rank.
top_nOPTIONAL- The maximum number of ranked documents to return.
return_documentsOPTIONAL- Whether to include the ranked document text in the response.
PROVIDER-SPECIFIC WORKFLOWS
Jobs this Agent App can support
Inspect Create Ai Search Completion
Create a non-streaming AI search chat completion with Baidu Qianfan. Use this as a bounded discovery step, retain the returned identifier, and avoid expanding the read beyond the task's stated scope.
create_ai_search_completionControl Create Air Image Generation
Generate images with the Baidu Qianfan MuseSteamer Air image endpoint. Resolve the target first, present material changes for confirmation, and make retries idempotent where the provider supports it.
create_air_image_generationVerify with Create Air Image Generation
Generate images with the Baidu Qianfan MuseSteamer Air image endpoint. Compare the returned provider state with the intended outcome and preserve stable IDs or canonical links in the run record.
create_air_image_generationSAFETY BOUNDARY
Operate Baidu Qianfan with explicit limits
- Authorize Baidu Qianfan with API key and grant only the provider access required by the selected actions.
- 14 actions are change-capable by name. Confirm the target identity and material parameters before allowing a write.
- Treat `cancel_batch`, `cancel_video_generation_task`, and `delete_response` as high-impact operations and require an explicit approval boundary.
- No provider scope string is declared on the sampled actions. Verify the connected account's actual permissions in Baidu Qianfan rather than assuming unrestricted access.
- After a call, inspect the structured result and execution record before reporting that the Baidu Qianfan task completed successfully.
CONNECTION MODEL
API key
Baidu Qianfan API key used with the Authorization Bearer header. Create it in System Management > API Key: https://cloud.baidu.com/doc/qianfan/s/wmh8l6tnf
Working Machines stores provider credentials behind the execution boundary. An agent can use an authorized connection identity, but catalog discovery alone does not reveal OAuth tokens, API keys, or provider secrets.
- READ SIGNALS
- 10
- WRITE SIGNALS
- 11
- HIGH IMPACT
- 3
- REVIEW SIGNALS
- 1
EVIDENCE AND AVAILABILITY
Provider reference
Action names, input fields, authentication types, and counts on this page are generated from the Working Machines provider catalog. Provider behavior, quotas, object semantics, account eligibility, and regional availability remain governed by Baidu Qianfan.