wave-update

Use this skill at the end of every wave, after `wave-execute` has produced an execution report. The skill reviews the wave (via a fresh-context subagent), surfaces findings to the human for approval, applies any architectural changes, closes the current wave, and expands the next wave's sketch into full detail. Triggers include phrases like "update the wave plan after W<N>", "close out W<N> and plan W<N+1>", "run the review and update", "what comes next after this wave's execution", or whenever an execution report exists for the current wave and the user is ready to advance the cycle. Also trigger when a vision pivot has just happened and the user wants the wave plan reconciled. Do NOT use when no execution report exists yet (run wave-execute first). Do NOT use to write code. Do NOT use mid-wave; this skill operates at wave boundaries only.

Safety Notice

This listing is imported from SkillsMP metadata and should be treated as untrusted until upstream source review is completed.

Copy this and send it to your AI assistant to learn

Install skill "wave-update" with this command: npx skills add vtphan/skillsmp-vtphan-vtphan-wave-update

No markdown body

This source entry does not include full markdown content beyond metadata.

Source Transparency

This detail page is rendered from real SKILL.md content. Trust labels are metadata-based hints, not a safety guarantee.

Related Skills

Related by shared tags or category signals.

Coding

design

Use this ACCORD skill when an approved intent exists and the human lead wants to codesign or revise architecture, boundaries, project commands, verification expectations, and consequential decisions. Triggers include "accord design", "draft the design", "revise architecture", "pivot design", or when implementation learning requires a new design version. This skill uses monotonic draft rounds, promotes an approved draft to docs/accord/design/design.md, updates accord-state.md, then commits and tags after human approval.

Repository SourceNeeds Review
Coding

execute

Use this ACCORD skill when canonical intent, design, and plan artifacts exist and the human lead wants the current approved unit implemented. Triggers include "accord execute", "execute the current unit", "implement the approved unit", "build the next unit", or "go ahead" after an approved ACCORD plan. This skill writes code, writes docs/accord/reports/exec-<unit-id>.md using the approved u-NNN-slug ID, updates accord-state.md, then commits and tags after human approval.

Repository SourceNeeds Review
Coding

intent

Use this ACCORD skill when the human lead wants to codesign or revise project intent before architecture/design work. Triggers include "accord intent", "draft project intent", "revise intent", "pivot intent", or when a project needs a lightweight approved intent artifact for downstream design and planning. This skill uses monotonic draft rounds, promotes an approved draft to docs/accord/intent/intent.md, updates accord-state.md, then commits and tags after human approval.

Repository SourceNeeds Review
Coding

architect

Use this skill at two distinct moments in a VADER project. (Mode = draft) When a vision doc exists and an initial architecture doc is needed before drafting the wave plan. Triggers include phrases like "draft the architecture for this project", "produce an architecture doc and Decision Log from this vision", "design the modules and key interfaces for X", or whenever the user wants the structural decisions for a new project committed to writing before wave planning starts. (Mode = ratify) After the user has reviewed the proposed architecture and decisions, to flip Proposed Decision Log entries to Accepted in one pass. Trigger when the user says "ratify the architecture" or "accept the proposed decisions". Do NOT use for mid-cycle architectural revision — that's handled by `wave-update`'s review subagent. Do NOT use to produce code, scaffolding, or implementation; this skill produces decision documents only.

Repository SourceNeeds Review
wave-update | V50.AI