Skip to content

Inventory: Expand, flatten, and curate outline information elements #24

@amotl

Description

@amotl

Problem

An average MCP documentation server subsystem can't easily follow nested resources recursively, specifically if the content markup does not convey full URLs to be fetched subsequently. For example, the resource about CrateDB reference documentation's index page clearly demonstrates the situation.

There is a little innocent-looking backlog item, which is a leftover work item from the initial sweep of distilling a rough outline, where this obstacle has not been considered yet on a few spots.

about/docs/backlog.md

Lines 7 to 9 in 7739eaf

- Outline: Revisit `notes` in `cratedb-outline.yaml`, for example about
pulling in `ctk docs functions` and `ctk docs settings`, but also about
other resources that have not been expanded properly yet

Details

The problem refers to the spots in cratedb-outline.yaml outlined below. It needs additional work in the spirit of curating ingredients optimally for this use case.

Tip

You can quickly discover is is not enough to enumerate high-level entrypoints of corresponding subsections of the knowledge tree when following corresponding link references: We need to dig deeper, in order to select, expand, and flatten relevant nested elements.

Enumerated spots

- title: "CrateDB reference documentation"
link: https://cratedb.com/docs/crate/reference/en/latest/_sources/index.rst.txt
description: The reference documentation of CrateDB.
notes: This just includes the index, but no content yet. Expand!
- title: "CrateDB Cloud"
link: https://cratedb.com/docs/cloud/en/latest/_sources/index.md.txt
description: The full documentation for CrateDB Cloud.
notes: This also just includes the index, but no content yet. Expand!

- title: "CrateDB drivers"
link: https://cratedb.com/docs/crate/clients-tools/en/latest/_sources/connect/index.md.txt
description: How to connect to a CrateDB cluster using traditional database drivers.
notes: This just refers to the index page, and does not include details about how to connect with individual drivers.

- title: "Time Series with CrateDB"
link: https://github.com/crate/cratedb-examples/raw/refs/heads/main/topic/timeseries/README.md
description: Examples, tutorials and runnable code on how to use CrateDB for time-series use cases. Exploratory data analysis, time-series decomposition, anomaly detection, forecasting.
notes: That's just the index again, and doesn't include any ingredients.
source: examples

- title: "Integration Tutorials I"
link: https://cratedb.com/docs/guide/_sources/integrate/index.md.txt
description: Integrating 3rd party software with CrateDB.
source: tutorials
- title: "Integration Tutorials II"
link: https://community.cratedb.com/raw/1015/1
description: Overview of CrateDB integration tutorials.
source: tutorials

Metadata

Metadata

Assignees

No one assigned

    Labels

    documentationImprovements or additions to documentation

    Type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions