From a7b7b154250e66da4cab80ab476573e781567e38 Mon Sep 17 00:00:00 2001 From: Connor Rhodes Date: Tue, 12 May 2026 12:51:28 -0500 Subject: [PATCH] pov-doc: add MongoDB dictations as data source for account context --- pov-doc/SKILL.md | 2 ++ 1 file changed, 2 insertions(+) diff --git a/pov-doc/SKILL.md b/pov-doc/SKILL.md index 9fbd1a7..cf612af 100644 --- a/pov-doc/SKILL.md +++ b/pov-doc/SKILL.md @@ -23,6 +23,8 @@ Before generating the document, gather the following from the user's notes and c Search `~/notes` for account notes, meeting notes, and inbox items related to the customer to populate these fields. Use ripgrep to find relevant content across all note files. +Additionally, query the MongoDB `notes.dictations` collection for meeting transcripts related to the customer. Transcripts often contain richer detail than summarized notes, including exact stakeholder names, technical requirements, and pain points discussed during calls and on-site visits. Search across the `content`, `title`, and `summary` fields using case-insensitive regex matching on the customer name and known contacts. The MongoDB skill provides connection details and query patterns. + ### Logo and Mission Statement In addition to the fields above, gather these for the document header: