Use DevonAgent to Search in DevonThink Pro Office

I have both products installed, DTProOffice and DevonAgent.

Maybe I am too dumb to find it out, but what I am trying to do is to search my DTPro Database for specific keywords like from a Contract eg “Supplier NEAR Termination”. That works insofar as DT displays a huge amount of documents and marks up the search keywords. But when I open the document, I’d like to have the possibility to search for the same string further down in this special document (eg. go to next result), as it may occur at several points. But when I open the document in DT’s own preview I can search but I cannot search with the same parameters like using NEAR or BEFORE.

So I tried to search my DTPro Database with DA selecting only the module “DTPro” in the search sets. But it displays no results at all. Gorgeous would be if DA comes up with a list of documents like in the digest view and summarizes the relevant paragraphs… is that possible?

Thanks in advance for your help

You have to start the web server (see Preferences > Server) and enable sharing of the database(s) (see File > Database Properties) first.

Thanks Christian,

Ok didn’t think about the web-server - but now it is clear…

in principle it works now, however I experience a lot of crashes of DA… is there a reason for that? I further notice that it doesn’t find PDF although I have checked them in the documents to display…

any more hints from your side?



ok - regarding some missing PDFs I found out in the protocol of the search that some documents were classified as too big. where can I set limitations for filesize or is this something DA manages internally?

The max. size can be defined via Preferences > Search. However, could you please send the crash logs (see ~/Library/Logs/DiagnosticReports) to cgrunenberg - at - Thanks in advance!


I will forward the crashlogs to you in the course of the day.
In the preferences section however I see that the max size is limited to 4 MB. That’s ok for websearch but it doesn’t work for larger pdfs. Any way around that?


Not yet. How large are your PDF documents (MB and number of pages)?

well that depends… normal pdfs are small but once I run them through DTP’s OCR engine they inflate to 64 MB or so… (ok that was a very large report and not standard…) No of Pages +100

The next release will support up to 8 MB (as huge files containing lots of words/pages can slow down the digest a lot).

Thanks Christian,

I am just beginning to figure out the OCR-capabilities. In the meantime I have been able to skim down the 64 GB document to 24 MB just by setting other OCR preferences (150 dpi “same as original” not ticked). Currently processing on 50% quality… However I cannot imagine that I will get below 4…


One workaround for now might be to split the document into multiple ones (and to keep the original one too).