Operator Briefing

Operator Briefing is the per-repository briefing surface for operators. The plugin is off by default.

When enabled, it adds a top-level /briefing sidebar page, records queryable review findings, notable workflow-change facts, and early "blocked on you" signals from open design-doc discussion threads, then synthesizes stored narrative, chart, image, artifact, and link_card blocks into per-repository briefing revisions. Image and artifact blocks render existing typed artifacts captured by the summarized Workflow; the briefing does not generate fresh artifacts for the main page.

Scheduled generation is skipped when nothing changed in a repository since its last closed briefing. Manual regeneration always runs.

The plugin also records personalization in Agent Memory. Source preferences default to every source enabled; AI-created source changes are pending suggestions until the operator confirms them. Explicit briefing feedback writes a global user_pref memory, and completed dives use the same memory path with higher confidence. Briefing generation reads the effective source preferences and the operator's Agent Memory context before producing a revision.

Briefing narratives can seed a small set of wiki-style dive spans. Operators can also select arbitrary briefing text and click "More info"; both paths start a briefing_dive follow-up Workflow on the same live Briefing Job and write or revise a durable repository-scoped topic page.