Self-hosted knowledge management for SMEs: a split-screen Markdown editor whose sections an LLM refines while you write, and RAG question answering over the documents that result. FastAPI + Postgres/pgvector on the back, SvelteKit on the front, everything OpenAI-compatible and self-hostable. Squashed into a single commit; the development history stays local. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01CA43ZJda8Rbp2hKXNy8f6b
48 lines
1.4 KiB
YAML
48 lines
1.4 KiB
YAML
# Catalog blueprint — inert until an admin adds it (docs/authoring-templates.md).
|
|
# The English variant of `projekt-debrief` (same id, locale: en).
|
|
#
|
|
# Not in the starter set. Kept short on purpose: a debrief that asks for a
|
|
# full project report gets written by nobody.
|
|
id: projekt-debrief
|
|
name: "Project review: what we learned"
|
|
version: "1.0"
|
|
kind: authoring
|
|
locale: en
|
|
description: >
|
|
After a project, record what came of it and what should go differently
|
|
next time.
|
|
|
|
model:
|
|
temperature: 0.4
|
|
min_class_hint: "12b"
|
|
|
|
persona: |
|
|
You are a precise technical editor. After a project you record what it was
|
|
about and what there is to learn from it — plainly, in complete sentences,
|
|
without blame and without gloss. Numbers, dates and names stay exactly as
|
|
given. No filler, no emoji.
|
|
|
|
title_template: "Project review from {{date}}"
|
|
|
|
skeleton: |
|
|
## What it was about
|
|
|
|
## How it went
|
|
|
|
## What we do differently next time
|
|
|
|
sections:
|
|
- heading: "What it was about"
|
|
hint: >
|
|
Starting point and goal of the project, in two or three sentences,
|
|
with the timeframe and who was involved.
|
|
- heading: "How it went"
|
|
hint: >
|
|
The outcome and the road to it — what worked and where it snagged.
|
|
- heading: "What we do differently next time"
|
|
hint: >
|
|
The concrete conclusions: what to repeat and what to avoid.
|
|
|
|
metadata:
|
|
visibility: department
|