openai deep research

Async deep web research powered by OpenAI. Tasks typically take 10–30 minutes; results are saved to file storage.

Server path: /embedded-openai-deep-research | Type: Embedded | PCID required: No

Tools

Tool	Description
embedded-openai-deep-research_check_status	Check the status of a previously started deep research task. Poll this until status is “completed” or “failed”. When completed, the final results are also saved to the file storage collection from the original request.
embedded-openai-deep-research_research	Start an async deep research task using OpenAI deep research models (10–30 minutes). Saves the result to a file storage collection — defaults to “Multimedia Artifact” unless collectionId is set. Returns a responseId and signedUrl to track progress.

---

embedded-openai-deep-research_check_status

Check the status of a previously started deep research task. Poll this until status is “completed” or “failed”. When completed, the final results are also saved to the file storage collection from the original request.

Parameters:

Parameter	Type	Required	Default	Description
responseId	string	Yes	—	The responseId returned from embedded-openai-deep-research_research

Show inputSchema

{
  "type": "object",
  "properties": {
    "responseId": {
      "type": "string",
      "description": "The responseId returned from embedded-openai-deep-research_research"
    }
  },
  "required": [
    "responseId"
  ]
}

---

embedded-openai-deep-research_research

Start an async deep research task using OpenAI deep research models (10–30 minutes). Saves the result to a file storage collection — defaults to “Multimedia Artifact” unless collectionId is set. Returns a responseId and signedUrl to track progress.

Parameters:

Parameter	Type	Required	Default	Description
model	string	No	"o4-mini-deep-research"	Select the appropriate deep research model: - o4-mini-deep-research: Fast, cost-effective deep research (default) - o3-deep-research: Advanced reasoning, deep research tasks
userPrompt	string	Yes	—	The research question or task to investigate
fileUrls	string[]	No	—	Optional array of file URLs to include as research context
outputFileName	string	No	—	Optional output filename. Defaults to “openai-deep-research-result-<responseId>.txt”.
reasoningEffort	string	No	"low"	Reasoning effort level — lower effort is faster but less thorough
collectionId	string	No	—	Optional file storage collection ID to store the output. Defaults to the Multimedia Artifact collection.
triggerUrls	object	No	—	Called when research completes. Example: { “https://triggers.app.pinkfish.ai/ext/triggers/EXAMPLE_TRIGGER_ID”: “EXAMPLE_API_KEY” }

Show inputSchema

{
  "type": "object",
  "properties": {
    "model": {
      "type": "string",
      "enum": [
        "o3-deep-research",
        "o4-mini-deep-research"
      ],
      "description": "\n    Select the appropriate deep research model:\n    - o4-mini-deep-research: Fast, cost-effective deep research (default)\n    - o3-deep-research: Advanced reasoning, deep research tasks\n  ",
      "default": "o4-mini-deep-research"
    },
    "userPrompt": {
      "type": "string",
      "description": "The research question or task to investigate"
    },
    "fileUrls": {
      "type": "array",
      "items": {
        "type": "string"
      },
      "description": "Optional array of file URLs to include as research context"
    },
    "outputFileName": {
      "type": "string",
      "description": "Optional output filename. Defaults to \"openai-deep-research-result-<responseId>.txt\"."
    },
    "reasoningEffort": {
      "type": "string",
      "enum": [
        "minimal",
        "low",
        "medium",
        "high"
      ],
      "default": "low",
      "description": "Reasoning effort level — lower effort is faster but less thorough"
    },
    "collectionId": {
      "type": "string",
      "description": "Optional file storage collection ID to store the output. Defaults to the Multimedia Artifact collection."
    },
    "triggerUrls": {
      "type": "object",
      "additionalProperties": true,
      "description": "Called when research completes. Example: { \"https://triggers.app.pinkfish.ai/ext/triggers/EXAMPLE_TRIGGER_ID\": \"EXAMPLE_API_KEY\" }"
    }
  },
  "required": [
    "userPrompt"
  ]
}