How to find out what a template needs and why it failed Leave feedback

Run the GroupDocs.Assembly MCP server locally and your agent can tell you what a template needs before anything is generated, and explain in plain words why it failed afterwards — a placeholder in the wrong syntax, an unclosed block, a missing field:

dnx GroupDocs.Assembly.Mcp --yes

What does order-lines.docx need, and why won’t it generate?

Note

The commands and config snippets on this page are for the .NET build of the server — the only platform available today. Installation and client setup: MCP server for .NET. Other platforms will expose the same tools with their own launch command; everything else on this page applies unchanged.

• The server registered in your AI client — quick start.

get_template_fields  -- static: reads the tags, generates nothing
   |   tags, dataSourceNames, fields, dataSkeleton, notes
   |   notes flag: {{...}} placeholders, unreadable binary templates
   v
assemble_document    -- the engine: finds what static reading cannot
   |   unclosed <<foreach>> ...... "Unexpected end of template"
   |   field missing in data ..... "The template reads 'Email', ..."
   |   empty list + arithmetic ... "Can not apply operator '*' ..."
   v
assemble_document {"inlineErrors":true}
       writes "Error! ..." into the output next to each bad tag

Captured from server 26.11.0 (the evaluation note that opens each response is trimmed).

A template in another engine’s syntax (Dear {{name}}, your order {{order_id}} has shipped.):

-> get_template_fields {"template":{"filePath":"shipping-notice.docx"}}
<- { "tagCount": 0, "notes": [
     "No <<…>> tags were found: this document is not a GroupDocs.Assembly
      template, or its tags are in a part the server does not read (text
      boxes in charts, embedded objects).",
     "The template contains 2 '{{…}}' placeholder(s). GroupDocs.Assembly
      does not process that syntax and leaves it as text; write <<[name]>>
      instead." ] }

An unclosed block — get_template_fields lists the tags (it reads them statically); the engine finds the problem:

-> assemble_document {"template":{"filePath":"order-lines.docx"},
     "dataJson":"{\"lines\":[{\"Title\":\"Cable\",\"Quantity\":2}]}"}
<- Document assembly failed for 'order-lines.docx': ... Unexpected end of
   template. A block tag is not closed: every <<foreach>> needs <</foreach>>
   and every <<if>> needs <</if>>.

Errors written into the output — inlineErrors: true generates anyway and marks each bad tag (here the upstream sample with << var [name]>>):

-> assemble_document {"template":{"filePath":"inline-error.docx"},
     "dataJson":"{\"name\":\"x\"}","inlineErrors":true}
<- Assembled 'inline-error.docx' with inline JSON (JSON object as 'product')
   as DOCX. The template has errors: each one is written into the output
   next to its tag, starting with 'Error!'

The document then reads << var [name] Error! An assignment operator is expected. >>.

A binary template:

-> get_template_fields {"template":{"filePath":"contracts-table.msg"}}
<- { "readable": false, "notes": [ "Tags cannot be read from a .msg template
     without generating it. Save the template as DOCX, XLSX, PPTX or EML to
     inspect it, or generate it with assemble_document and inlineErrors=true
     to see which fields fail." ] }

Markdown is a template too — product-readme.md reported "dataSourceNames": [ "product", "description" ] and generated a DOCX.

Before a run: the list of data the template needs, and any syntax the engine will ignore. After a failed run: the reason in one sentence and the fix.

Is the skeleton exact? It is inferred from the tags; the server says so in notes. Values computed through Select or GroupBy projections are not followed.

Why do tags in a chart not show up? Static reading skips some parts (text boxes in charts, embedded objects); the note says so. The engine still processes chart tags — see reports with charts.

Where is the syntax documented? In the library: template syntax, part 1 and part 2, and inline error messages.

Agents follow project instructions about when to use a tool. Paste this into the instruction file your agent reads (CLAUDE.md, AGENTS.md, .github/copilot-instructions.md, .cursorrules):

Before filling a template, call get_template_fields and read its notes. If
assemble_document fails, quote the server's message, fix the data or tell me
the template line to change; use inlineErrors=true only for debugging copies.

• Evaluation mode: every generated document starts with an Evaluation Only line, workbooks gain a worksheet, and e-mail templates lose their HTML body. Licensing.
• <<[name]>>, not {{name}}: other engines’ placeholders stay in the output as text.
• Templates are trusted input: image and document tags load the paths and URLs their expressions return.
• Dates: write them as "2026-11-12T00:00:00" and format them in the template; a date-only string can shift by a day on machines east of UTC.

• Generate documents with AI agents
• One document per record (mail merge)
• Reports with tables and charts
• From notes or email to a document
• Reports from repository data
• On-premise deployment
• Tools reference · Licensing · Troubleshooting & FAQ
• Questions: Assembly forum

• What does this template need? Ask it. — the same scenario on the GroupDocs blog, with worked examples and the trade-offs.