dev-workflow 2.0.0

Major release. Breaking: the explain-in skill is removed.

Removed

• explain-in — the leadership-reframing skill. Any workflow that invokes it by name must be updated. post-mortem no longer offers the leadership handoff. The name is reused by explain-in-html, which does something different.

Added

• explain-in-html — renders output as an HTML page instead of terminal text. Two modes inferred from content:

  • explain — Mermaid, UML, ER and state charts; hand-authored inline SVG for before/after and side-by-side comparisons
  • interact — a question round answered in-browser, persisted to the artifact server via db and read back with ArtifactData

  It owns the medium, not the decision: grilling, brainstorming and scrutinize keep deciding what to ask or say, run first, and hand the rendering here. Figure work is delegated to the built-in artifact-diagramming (mechanism) and dataviz (quantities) skills rather than restated. Falls back to a local docx/explainers/<slug>.html, and to AskUserQuestion in interact mode.

• docx/explainers/ added to the storage convention.

Fixed

• Long-stale skill counts in dev-workflow/README.md ("thirteen"/"13") and PLUGINS_INSTALLATION.md ("20").

---

24 skills, 8 agents, 7 commands, 4 templates.