eng-design-doc-reviewlisted

# Engineering Design Doc Review — Independent Fresh-Context Audit Adversarially review a design document with fresh context. The brief this skill dispatches lives in `skills/reviewing-designs/SKILL.md`, and the orchestrator loads the same brief for the DESIGN phase's adversarial review gate. Invoking this skill standalone remains supported whenever you want an independent, fresh-context audit of a design document. Write the prose this skill governs at a seventh-grade reading level, in STE-flavored mode — short sentences, common words, no unexplained jargon. Full methodology: `writing-prose`. Before you finalize prose this skill governs, call the Skill tool with `writing-prose` and apply its `## Self-lint` checklist. There is **no custom review agent**. This skill loads the review brief from `reviewing-designs` and dispatches the built-in read-only `Explore` subagent through the `Agent` tool. That subagent boots with a **clean context** and no shared conversation history with the design-author — that isolation is the whole point. It prevents self-evaluation bias. `Explore` holds no Write/Edit tools, so the reviewer structurally cannot change the artifacts it judges. Fresh context plus veto-without-authorship is the generator-evaluator rule (`skills/principle-generator-evaluator/SKILL.md`). ## Input `$ARGUMENTS` is the artifact directory: `docs/plans/<id>/`. If empty, the discovery block below resolves it. The review reads: - `$ARGUMENTS/design.md` — the document under r