菜单
Fetch npm package updates, analyze new features, and identify improvement opportunities for this app. Use when the user wants to explore what's new in dependencies or plan next development directions.
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
Run a Vibeman pipeline on a project. Three modes — (A) goal-based development (PLAN > IMPLEMENT > VERIFY > REPORT), (B) audit-driven scan + triage + wave-based fix implementation, or (C) scan-and-decide (pick one context group; the skill auto-selects Idea scanners, generates a capped backlog, you accept/reject each, then it implements the approved scope). Scans are memory-backed — a per-project Obsidian vault records which contexts were scanned with which lens at which commit, what was fixed, and what the user rejected, so repeat runs target new ground instead of restarting from scratch. Read-only modes — status, coverage, reflect. All run with quality gates, brain-signal recording, and structured per-wave/per-phase reporting.
Session-after-session product perfection loop for Vibeman. The strongest available model (Fable) directs — it walks the repo's context map context-by-context, proposes 5 challenged, high-value directions per context (features, design elevations, significant optimizations), gates them with the user until 10 are accepted, then orchestrates one Opus builder subagent per context in isolated worktrees while making every review/merge decision itself. All state lives in a linked Obsidian vault so any future session resumes the loop exactly where the last one stopped. Invoke with `/perfect [init|propose|build|status|reflect] [context-name]`.
Systematically triage, challenge, and process a backlog of auto-generated idea files. Groups by context area, validates against codebase, filters BS, presents one approval gate, then executes autonomously with code review.
Generate images with Leonardo AI (Lucid Origin model), remove backgrounds, and write SVG. For brand assets, UI illustrations, backgrounds, and icons.
Review code for quality, security, and patterns. Scans changed files for bugs, OWASP vulnerabilities, SQLite injection, missing error handling, and anti-patterns specific to this Next.js + SQLite + Zustand stack.
Test-driven development workflow. Write failing tests first, then implement minimal code to pass, then refactor. Uses Vitest with better-sqlite3 test database. Enforces RED → GREEN → IMPROVE cycle.
| name | npm-updates |
| description | Fetch npm package updates, analyze new features, and identify improvement opportunities for this app. Use when the user wants to explore what's new in dependencies or plan next development directions. |
| disable-model-invocation | true |
| allowed-tools | Read, Grep, Glob, Bash(npm outdated*), Bash(npm view*), Bash(npm info*), WebSearch, WebFetch |
| argument-hint | ["category?"] |
Analyze npm dependency updates and identify new features that could improve this app.
If $ARGUMENTS is provided, filter analysis to that category only (e.g. "ai", "ui", "framework", "data", "testing").
Run npm outdated --json in the project root to get a structured list of packages with available updates. Parse the JSON output to identify:
Focus analysis on major and minor updates. Patches can be listed but don't need deep analysis.
Group the outdated packages by their role in the app:
For each package with a major or minor version bump:
<package-name> release notes <latest-version>)Cross-reference new package features with the app's architecture:
src/app/features/Manager/lib/conductor/): Could AI SDK updates improve the pipeline stages?src/lib/claude-terminal/, src/components/cli/): New Claude SDK features?src/app/features/TaskRunner/): Better state management, UI improvements?@xyflow/react): New node types, interaction patterns?src/app/db/repositories/): Better query patterns, perf improvements?src/lib/supabase/): Supabase SDK improvements?src/mcp-server/): MCP SDK protocol updates?Output a structured report:
Package Update Summary as a markdown table with columns: Package, Current, Latest, Update Type, Priority.
Improvement Opportunities ranked by impact, each with:
Recommended Update Order:
Breaking Changes to Watch - list any breaking changes requiring migration work.
After presenting the report, ask the user which improvement opportunities they want to pursue. Offer to: