Data Flywheel

BeeWeave is designed as a loop, not a one-way archive.

collect -> create -> distill -> reuse context -> collect better material -> create the next piece

1. Collect

Put raw material into workbench/: links, notes, PDFs, exported chats, meeting transcripts, screenshots, product briefs, source files, and findings from agent sessions.

/beeweave-ingest workbench/inbox

2. Create

Use the workbench as the drafting surface. Writing and workbench skills can turn collected material into articles, social posts, research notes, specs, or project output while keeping the raw sources nearby.

3. Distill

After work is published or a decision becomes stable, ingest the high-signal output into the vault:

/beeweave-article-publisher workbench/articles/drafts/my-article.md
/beeweave-ingest workbench/articles/published
/beeweave-update

Use beeweave-article-publisher for finished article drafts: it moves one draft to workbench/articles/published/, marks it published, and ingests that file into the wiki. Use beeweave-ingest directly for folders or non-article source material.

The vault should hold durable concepts, entities, references, synthesis notes, project decisions, and linked Markdown pages.

4. Reuse Context

Before the next task, query the vault so the agent starts from what you already know:

/beeweave-query what do I know about MCP security?

5. Collect Better and Create the Next Piece

Queries reveal gaps, drafts expose weak claims, and digests surface emerging themes. Use those signals to decide what to collect next and what to create from it.