‹ 首页

dev-ralplan

@evolution-foundation · 收录于 5 天前 · 上游提交 2 个月前

Multi-perspective consensus planning with Planner/Architect/Critic loop. Use when high-stakes decisions need RALPLAN-DR structured deliberation (auth, migrations, public APIs, irreversible changes). Pairs with @compass-planner + @apex-architect + @raven-critic.

适合你,如果需要在团队中结构化地讨论和决策重大技术变更。

/ 通过 npx 安装 校验哈希
npx oh-my-skill add evolution-foundation/evo-nexus/dev-ralplan
/ 通过 bash 安装
curl -fsSL https://oh-my-skill.com/install.sh | bash -s -- evolution-foundation/evo-nexus/dev-ralplan
/ 已经装过?验证本机副本,不用重装
npx oh-my-skill verify evolution-foundation/evo-nexus/dev-ralplan
安装目标可用 --agent / --scope 或 --to 明确指定;省略时只会在唯一已存在的 agent 目录上自动选择,零命中或多命中会停止并提示。content_hash 缺失或不一致均拒装。
509GitHub stars
~547上下文体积 · 单文件
索引托管

怎么用

商店整理自技能原文 · 版本 7f5dd76 · 表述以原文为准
它做什么

装上后,Claude会启动一个由规划师、架构师和评论家角色组成的多人共识规划流程,生成带有原则、决策驱动因素、可行选项和架构决策记录的计划。

什么时候触发

当需要多视角共识来制定高风险或不可逆的决策(如身份验证、数据迁移、破坏性变更)时触发。你也可以直接说“ralplan”、“consensus”等关键词来激活。

装好后可以这样说
启动共识规划流程。
触发ralplan模式。
进入多角色审议。
技能原文 SKILL.md作者撰写 · Apache-2.0 · 7f5dd76

Dev Ralplan

Derived from oh-my-claudecode (MIT, Yeachan Heo). Adapted for the EvoNexus Engineering Layer.

Consensus mode planning: a Planner/Architect/Critic loop that produces a plan with explicit principles, decision drivers, viable options, and an ADR. Used for high-stakes or irreversible work.

Use When
  • Multi-perspective consensus needed on a plan or design
  • High-stakes work: auth, security, data migrations, destructive/irreversible changes, production incidents, compliance/PII, public API breakage
  • User says "ralplan", "consensus", "let's deliberate"
  • When dev-plan --consensus would normally apply
Do Not Use When
  • Simple, low-risk task → use dev-plan standard mode
  • User wants direct execution → use dev-autopilot
  • Single-perspective is sufficient → use dev-plan direct mode
Workflow
Stage 1 — Initial plan (@compass-planner)
  • Run interview if needed
  • Emit RALPLAN-DR summary: Principles (3-5), Decision Drivers (top 3), Viable options (≥2)
  • If only 1 viable option remains, document explicit invalidation rationale for alternatives
Stage 2 — Architecture review (@apex-architect)
  • Read the plan
  • Provide consensus addendum: strongest steelman antithesis, real tradeoff tension, synthesis if viable
  • In deliberate mode, flag explicit principle violations
Stage 3 — Critique (@raven-critic)
  • Multi-perspective adversarial review
  • Pre-commitment predictions
  • Gap analysis
  • Self-audit + realist check
  • Verdict: REJECT / REVISE / ACCEPT-WITH-RESERVATIONS / ACCEPT
Stage 4 — Revised plan (@compass-planner)
  • Incorporate apex + raven feedback
  • Final plan must include ADR: Decision, Drivers, Alternatives considered, Why chosen, Consequences, Follow-ups
  • In deliberate mode: pre-mortem (3 scenarios) + expanded test plan (unit/integration/e2e/observability)
Mode Selection
  • Short mode (default): principles + drivers + ADR
  • Deliberate mode (--deliberate): + pre-mortem + expanded test plan + principle violation flags

Trigger deliberate mode when: auth/security touched, data migration, destructive/irreversible change, production incident response, compliance/PII implications, public API breakage.

Output

Final plan saved to workspace/development/plans/[C]ralplan-{name}-{date}.md with ADR section.

Pairs With
  • @compass-planner (drives the loop)
  • @apex-architect (architecture review)
  • @raven-critic (adversarial review)
  • dev-plan (which can chain into ralplan with --consensus)
  • dev-autopilot (skips its own Phase 0+1 if a ralplan plan already exists)
按 Apache-2.0 许可原样转载,未经改动 · 在 GitHub 查看 →

评论

登录即可评论;带「已验证安装」的,是发布者名下有本店的安装或持有记录。