Forum Discussion

Jaimin26's avatar
Jaimin26
Copper Contributor
Aug 13, 2026

Copilot Studio: “Get file content using path” fails for table-heavy DOCX files

Hi everyone,

I'm investigating an issue with a Copilot Studio Standard Harness agent and would like to know whether others have encountered similar behavior.

Scenario

I have configured a tool in my agent using:

Get file content using path

The document is stored in my own OneDrive/SharePoint location and I have full access to it.

This is not a chat file upload scenario and not a knowledge source indexing scenario. The agent is retrieving the document through a configured tool action.

Behavior Observed

Case 1 - Works

The Word document contains approximately:

  • ~4.1K words
  • ~20 tables

The agent successfully:

  • Finds the file
  • Retrieves the file
  • Reads the content
  • Generates a summary

Case 2 - Fails

I increase the document slightly:

  • ~4.2K words
  • More tables added

The agent now returns a message similar to:

The file was retrieved, but its content could not be converted to text.

The response still indicates that:

  • File found 
  • File retrieved 
  • Text extraction/conversion failed 

Important Observation

Initially I thought this might be a word-count limitation, but additional testing suggests otherwise.

I tested another document containing:

  • More than 4.3K words
  • Very few or no tables

The agent processed that file successfully and generated a correct summary.

Because of this, the issue appears to be related more to table-heavy DOCX content than to the total number of words.

What Has Been Ruled Out

The following do not appear to be the cause:

  • File path issues
  • Permissions/access issues
  • File not found issues
  • Connector authentication issues
  • Large file size issues
  • Chat file upload limitations
  • Knowledge source indexing limitations

The same tool and same retrieval approach continue working until the document becomes more table-heavy.

Questions

  1. Has anyone experienced similar behavior with Get file content using path in Copilot Studio?
  2. Does Copilot Studio Standard Harness perform an internal DOCX-to-text conversion step after retrieving the file?
  3. Are there known limitations around processing Word documents that contain a large number of tables?
  4. Has anyone identified practical thresholds around table count, table complexity, or table-heavy technical documents?
  5. Is there a recommended workaround besides splitting the document into smaller files?

Any insights, similar experiences, or Microsoft guidance would be greatly appreciated.

Thanks!

ai agents
copilot studio
generative orchestration
microsoft 365 copilot

1 Reply

  • Jamony's avatar
    Jamony
    MCT
    Aug 15, 2026

    The test isolates the failure to text extraction after retrieval, not the path or authentication. The OneDrive action returns the file as binary content; Copilot performs the readable-text conversion afterward. Microsoft documents DOCX and file-size support, but publishes no table-count or complexity threshold. Make two copies of the working file and add one structural change at a time: merged cells, nested tables, text boxes, headers, and embedded objects. Word content in standard paragraphs and ordinary body tables is supported, while text in several other structures may be skipped. Simplify or flatten the first structure that reproduces the error, resave the document in desktop Word, and test again. For a reliable production workaround, pre-extract the content to plain text or HTML before passing it to the agent. If it still fails, submit both minimal files, timestamps, environment ID, and run IDs to Power Platform support.