菜单
Use when task involve design spec, ADR, one-way-door decision, or evaluate trade-offs between architectural approaches. Invoke before implementation when requirements span subsystems or decision hard reverse. Trigger: /architect.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Author, validate, and package grim-publishable artifacts — skill directories, rule files, agent definitions, MCP server descriptors, and bundle TOMLs. Use when creating or editing an artifact for grim build or grim release; when choosing frontmatter or catalog metadata fields; when adding claude, opencode, copilot, codex, cursor, kiro, junie, gemini, zed, or amp vendor keys; or when grim build fails validation with exit code 65.
Drive the grim CLI — the OCI package manager for AI skills, rules, agents, and bundles. Use when installing, updating, searching, or publishing AI-config artifacts with grim; when composing grim init, config, add, lock, install, update, status, context, fetch, describe, search, tui, mcp, build, release, publish, login, logout, or completions commands; when configuring settings, multiple registries, or qualified alias/repo references; or when resolving registries, project vs global scope, client targets, or offline mode.
Craft effective AI agent configuration — skills, rules, agents, and instructions — for any AI coding client. Use when writing or reviewing a SKILL.md, rule file, or agent definition; when deciding between a skill, rule, hook, or always-on instruction; when a config file grows past its context budget; or when a skill description fails to trigger.
Use for branch, PR, or diff review before landing on main. Tier (`low | auto | high | max`) scales breadth from single reviewer to adversarial panel with architect + SOTA-gap + Codex cross-model gate. Triggers: "review", "/swarm-review", pre-merge checks.
Use when the user says "codex review", "/codex-adversary", "adversarial review", "cross-model review", or asks for a different-model second opinion on a diff or plan artifact. Complements `/swarm-review` via Codex-CLI. Scopes: code-diff (default), plan-artifact (gate for `/swarm-plan`).
Use to implement a plan artifact from `/swarm-plan`, or a free-text implementation task with contract-first TDD + Review-Fix Loop. Tier (`low | auto | high | max`) scales builder model, loop rounds, review breadth, and Codex code-diff gate.
| name | architect |
| description | Use when task involve design spec, ADR, one-way-door decision, or evaluate trade-offs between architectural approaches. Invoke before implementation when requirements span subsystems or decision hard reverse. Trigger: /architect. |
Role: system design, technical specs, high-level architecture decisions for Grimoire.
worker-architecture-explorer. Map current module state, find reusable code, trace cross-module flowsworker-researcher for trending tools, proven patterns, industry adoption. Check/persist .claude/artifacts/research_*.md.claude/templates/artifacts/.claude/rules/arch-principles.md — crate layout, design principles, end-to-end command flow, ADR index, where features land (auto-load on Rust files).claude/rules/product-context.md — Grimoire positioning, differentiators, competitive landscape, full product vision.claude/rules/product-tech-strategy.md — golden-path tech choices.claude/rules/subsystem-*.md — per-subsystem invariants and module maps (load those match design area)mcp__context7__resolve-library-id + get-library-docs) — current crate API shape when design decision hinge on "what does crate X look like today". Training-data API knowledge decay fast. Fallback: WebFetch of docs.rs.mcp__github__*) — structured lookup of issues, PRs, releases during discovery. Fallback: gh CLI.Create in .claude/artifacts/ per CLAUDE.md naming:
adr_[topic].md — Architecture Decision Recordssystem_design_[component].md — system designsplan_[task].md — implementation plans$ARGUMENTS