菜单
Quick reference for all nanostack commands. Shows available skills, what each one does, and how to use them. Triggers on /nano-help.
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
Use when the user asks about available workflow skills, wants an overview of the engineering workflow, or references "nanostack". Also triggers on /nanostack.
Document what you learned during this sprint. Reads artifacts, writes structured solutions to know-how/solutions/. Run after /ship or after fixing a significant bug. Triggers on /compound.
Use after writing code to get a thorough code review. Runs two passes — structural correctness then adversarial edge-case hunting. Scales depth by diff size. Supports --quick, --standard, --thorough modes. Triggers on /review.
Use before shipping to production. Performs OWASP Top 10 audit and STRIDE threat modeling against the codebase. Supports --quick, --standard, --thorough modes. Also use when the user asks to check security, audit code, or review for vulnerabilities. Triggers on /security.
Use when you have a rough idea and want it sharpened before any code is written. Refines the idea through questions, one at a time. Explores 2-3 approaches with trade-offs. Walks the design with you, section by section. Saves a brief your next steps read. Still challenges scope and finds the smallest starting point. Supports --autopilot to run the full sprint automatically after approval. Use --retro after a sprint to reflect on what shipped. Triggers on /think, /office-hours, /ceo-review.
Use when working near production, sensitive systems, or destructive operations. Activates on-demand safety hooks that block dangerous commands. Supports modes — careful (warn), freeze (guided, keeps writes within scope), unfreeze (remove restrictions). Triggers on /guard, /careful, /freeze, /unfreeze.
| name | nano-help |
| description | Quick reference for all nanostack commands. Shows available skills, what each one does, and how to use them. Triggers on /nano-help. |
| concurrency | read |
| depends_on | [] |
| summary | Help and quick reference for nanostack skills. |
| estimated_tokens | 100 |
Show the user a concise overview of nanostack. No walls of text. Organized by what they want to do.
Defensive telemetry init. No-op if telemetry is disabled via NANOSTACK_NO_TELEMETRY=1, ~/.nanostack/.telemetry-disabled, or if the helpers are removed.
_P="$HOME/.claude/skills/nanostack/bin/lib/skill-preamble.sh"
[ -f "$_P" ] && . "$_P" nano-help
unset _P
Print this directly:
nanostack
Make your AI agent think first.
What do you want to do?
Start a brand new project → /think (or /think --autopilot)
Add to a project that exists → /feature <what to add>
Already wrote code, check it → /review (then /security, /qa)
Ready to save / publish → /ship
First time using nanostack → /nano-run
Not sure where to start → /think (it helps you decide)
Reference — every command:
Getting started:
/nano-run First-time setup. Configures your project conversationally.
/nano-help You are here.
The sprint:
/think Challenge the scope before building.
/nano Plan the implementation. Files, steps, risks.
build You or the agent writes the code.
/review Two-pass code review. Scope drift detection.
/security OWASP Top 10 + STRIDE audit. Graded A-F.
/qa Test it. Browser, API, CLI, or debug.
/ship Create PR, verify CI, generate sprint journal.
Shortcuts:
/think --autopilot Full sprint. Think, plan, build, review, audit, test, ship.
/feature <description> Add a feature to an existing project with full sprint.
After shipping:
/compound Save what you learned. Future sprints find it automatically.
Safety:
/guard Block dangerous commands. /freeze locks edits to a scope.
Team:
/conductor Parallel sprints across multiple agents or terminals.
Modes (for /review, /security, /qa):
--quick Small changes. Only the obvious.
--standard Default. Normal work.
--thorough Auth, payments, infra. Flag everything.
Update:
/nano-update Pull latest version.
/nano-doctor Diagnose install health (deps, perms, telemetry).
github.com/garagon/nanostack
If the user asks about a specific skill, invoke it: use Skill tool with the skill name. Don't explain the skill yourself — let the skill's own SKILL.md handle it.
Before returning control:
_F="$HOME/.claude/skills/nanostack/bin/lib/skill-finalize.sh"
[ -f "$_F" ] && . "$_F" nano-help success
unset _F
Pass abort or error instead of success if help did not complete normally.