nw-buddy
Read-only nWave concierge for methodology, current project evidence, command routing, migration, and troubleshooting.
Install / Use
npx skills add nWave-ai/nWave --skill nw-buddyInstalls into whichever agent you are using.
SKILL.md
Installable skill definition
Quality Score
Category
AI & Machine LearningSupported Platforms
Tags
Our assessment of nw-buddy
nw-buddy scores 81/100 on our quality scale, 734th of 950 AI & Machine Learning skills we index.
Its SKILL.md is 5.4 KB long, well organised into 10 sections with 1 code example: a solid amount of guidance for an agent.
It has 617 GitHub stars, a meaningful sign that others use it.
Maintenance, license and trust
- The repository was last updated about 2 months ago, so nw-buddy is actively maintained.
- Our last check on 2026-11-06 found the source still online.
- It is released under the MIT license, a permissive license that allows use, modification and commercial use with attribution.
- Its trust signals score 100/100, with no cautions. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.
Safety scan
No issues foundOur scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful.
AI review by kimi-k2.7-code on 2026-10-04. Automated pattern scan on 2026-10-04. It catches known dangerous patterns, not every risk — read a skill before letting an agent act on it.
nw-buddy compared with similar skills
All 4 of these similar skills score higher than nw-buddy; compare them before choosing.
| Skill | Score | Stars | Updated | Format |
|---|---|---|---|---|
| nw-buddy (this skill)by nWave-ai | 81 | 617 | 2mo ago | SKILL.md |
| claude-memby thedotmack | 100 | 117.9k | today | CLAUDE.md |
| Understand-Anythingby Egonex-AI | 100 | 93.0k | 2d ago | CLAUDE.md |
| headroomby headroomlabs-ai | 100 | 79.9k | today | CLAUDE.md |
| CowAgentby zhayujie | 100 | 48.0k | today | CLAUDE.md |
Frequently asked questions
- How do I install nw-buddy?
- Run
npx skills add nWave-ai/nWave --skill nw-buddy. The install tabs above show the steps for each supported agent. - Which AI agents does nw-buddy work with?
- It is written for Claude Code and OpenAI Codex, as a SKILL.md file. Other agents that read the same format can often use it too.
- Is nw-buddy safe to use?
- Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful. It is MIT-licensed and scores 100/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
- Is nw-buddy still maintained?
- The repository was last updated about 2 months ago, so nw-buddy is actively maintained.
Skill content
View source on GitHubname: nw-buddy description: "nWave concierge — ask any question about methodology, project state, commands, migration, or troubleshooting. Read-only, contextual answers." user-invocable: true argument-hint: '[question] - Example: "What should I do next for rate-limiting?"'
NW-BUDDY: nWave Concierge
Wave: CROSS_WAVE | Agent: Guide (nw-nwave-buddy) | Command: /nw-buddy
Overview
Ask questions about nWave — methodology, project state, commands, migration, troubleshooting. Guide reads your project and gives contextual answers.
Configuration questions — mandatory Read-tool instruction (D7)
Before answering ANY question about nWave configuration (rigor profile, documentation density, expansion mechanism, doc tiers, audit logging, etc.):
- Read
docs/reference/global-config.mdusing the Read tool — this is the canonical schema reference. Do NOT answer from training memory; configuration evolves and your training is not authoritative. - Read
docs/guides/configuring-doc-density.mdfor the how-to perspective if the question is about changing settings. - Then answer with grounded specifics: cite the exact config key path, valid values, and link to the source doc.
Why: configuration drift between releases would otherwise cause stale answers. The reference doc is the contract. (See ADR-001-rigor-as-config-concern.md for the rigor/density inheritance decision.)
Expansion mechanism — answering "how do I see more detail?"
When the user asks how to see more detail in their feature documentation (or equivalent — "expand", "more rationale", "why this decision", "show alternatives"):
- Explain the expand mechanism for wave commands: each wave (
/nw-discover,/nw-discuss,/nw-design,/nw-devops,/nw-distill,/nw-deliver) accepts--expand <id>to render a specific Tier-2 rationale section into the wave'sfeature-delta.mdwithout re-running the wave. - Mention the wave-end interactive prompt: when
documentation.expansion_prompt = "ask", each wave ends with a menu listing available expansions; the user picks items by id and the wave appends them tofeature-delta.md. - Provide example expansion identifiers — at minimum three across waves:
jtbd-narrative(DISCUSS),alternatives-considered(DISCUSS),migration-playbook(DISCUSS),trade-off-analysis(DESIGN),c4-narrative(DESIGN),infra-cost-analysis(DEVOPS),runbook-drafts(DEVOPS),edge-case-enumeration(DISTILL),refactoring-journal(DELIVER). Each wave's full expansion catalog lives in its skill's "Output Tiers" section.
Direct the user to docs/guides/configuring-doc-density.md for the persistent configuration path, and to the relevant wave skill (~/.claude/skills/nw-{wave}/SKILL.md) for that wave's expansion catalog.
Graceful degradation — when the reference is missing
If docs/reference/global-config.md is absent (Read tool returns not-found):
- State that the configuration reference document is unavailable — do NOT fabricate config keys or valid values.
- Direct the user to the troubleshooting path: check that nwave-ai is installed (
uv tool list 2>/dev/null | grep nwaveorpipx list 2>/dev/null | grep nwave), thatdocs/reference/exists in the project, or runpython -m nwave_ai.cli installto restore framework docs. - Offer to answer non-configuration questions while the reference is being restored.
Version-awareness — answering "what's new in vX.Y?" / "what changed?" / "what was fixed?"
When the user asks about release content, fixes, or improvements in a specific or current version:
- Read
docs/guides/whats-new-v<MAJOR><MINOR>/README.md— there is one folder per minor release (e.g.whats-new-v314/,whats-new-v35/). The pattern iswhats-new-v+ version digits without dots. - If the user asks about the current installed version, read
nWave/VERSIONfirst to determine which whats-new file to load. - If the user asks about a bug fix or symptom (e.g. "is uninstall residuals fixed?", "is pre-push faster?"), check
docs/guides/troubleshooting-guide/README.mdfirst — fix-recovery instructions live there alongside symptom descriptions. - Cite specific section anchors (e.g.
whats-new-v314/#improvements-in-v3140-rc1-2026-05-03) so users can deep-link. - If the user is on an older version and asks about upgrade impact, point them to the Upgrading from vN-1 section of the relevant whats-new file.
Why: release content evolves per RC/release; the whats-new files are the canonical changelog grain. Don't answer from training memory — versions move.
Agent Invocation
@nw-nwave-buddy
Execute *help to show capabilities, or ask any nWave question directly.
Configuration:
- model: sonnet (optimized for frequent, low-cost interactions)
- mode: read-only (never creates or modifies files)
Success Criteria
- [ ] Question answered with project-specific context (not generic docs)
- [ ] File paths verified against actual filesystem before citing
- [ ] Specialist agent/command recommended when deeper expertise needed
Examples
/nw-buddy Where are the documents for my rate-limiting feature?
/nw-buddy What should I do next?
/nw-buddy What's JTBD?
/nw-buddy How do I migrate to the SSOT model?
/nw-buddy My DISTILL is failing — architecture missing
Expected Outputs
- Conversational answers grounded in project state
- Command recommendations with rationale
- Feature progress dashboards when requested
Related Skills
claude-mem
117.9kPersistent Context Across Sessions for Every Agent – Captures everything your agent does during sessions, compresses it with AI, and injects relevant context back into future sessions. Works with Claude Code, OpenClaw, Codex, Gemini, Hermes, Copilot, OpenCode + More
Understand-Anything
93.0kGraphs that teach > graphs that impress. Turn any code into an interactive knowledge graph you can explore, search, and ask questions about. Works with Claude Code, Codex, Cursor, Copilot, Gemini CLI, and more.
headroom
79.9kCompress tool outputs, logs, files, and RAG chunks before they reach the LLM. 20% fewer tokens for coding agents, 60-95% fewer tokens for JSON, same answers. Library, proxy, MCP server.
CowAgent
48.0kOpen-source personal AI assistant & Agent Harness. Plans tasks, runs tools and skills, self-evolves with memory and knowledge. Multi-agent, multi-model, multi-channel. Lightweight, extensible, one-line install.
Languages
Trust signals
From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.
