Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/netwaif/multi-agent-starter/configure-multiagentnpx skills add netwaif/multi-agent-starter --skill configure-multiagentgit clone --depth 1 https://github.com/netwaif/multi-agent-starterWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/netwaif/multi-agent-starter/configure-multiagent)<a href="https://agentmods.dev/skills/netwaif/multi-agent-starter/configure-multiagent"><img src="https://agentmods.dev/badge/skills/netwaif/multi-agent-starter/configure-multiagent.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00118 | $0.01061 |
| Opus 5 | $0.00059 | $0.00531 |
| Sonnet 5 | $0.00024 | $0.00212 |
| Haiku 4.5 | $0.00012 | $0.00106 |
Grade A, and why
configure-multiagent scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 38 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Configure MultiAgent System
이 스킬은 file-as-memory 멀티에이전트 오케스트레이션 시스템을 대상 폴더에 생성한다.
직접 파일을 손으로 쓰지 말 것 — 반드시 이 스킬 폴더에 함께 들어있는 결정적 생성기 generator/init.py를 실행한다 (불변식·일관성 보장).
절차
- flavor 확인 — 사용자에게 묻는다(또는 현재 호스트로 제안):
claude— Claude Code 오케스트레이터 (워커: claude-main / codex-main / codex-critic / gemini)codex— Codex 오케스트레이터 (워커: codex-main / claude-critic / gemini)antigravity— Antigravity 오케스트레이터 (Gemini 3.1 Pro High; 워커: claude-main / codex-main / codex-critic, 멀티모달은 오케스트레이터 직접)
- 대상 폴더 확인 — 어디에 설치할지 묻는다. (상위 폴더 오인 주의 — 정확한 경로를 확인받는다.)
- 생성기 위치 —
generator/는 이 SKILL.md와 같은 폴더 안에 있다(스킬 자기완결). 호스트별 분기 불필요 — 이 스킬 폴더 기준./generator/init.py. (Claude는$CLAUDE_PLUGIN_ROOT/skills/configure-multiagent/generator/init.py로 해석됨.) - 실행 — 확인 후 (이 스킬 폴더의 generator 경로로):
대화형으로 진행하려면 인자 없이 실행하면 메뉴가 뜬다.python3 "<이 스킬 폴더>/generator/init.py" --flavor <claude|codex|antigravity> --target "<대상폴더>" --yes - 결과 보고 —
init.py가 끝에validate.py를 자동 실행한다. 그 PASS/FAIL을 그대로 사용자에게 보고한다. FAIL이 하나라도 있으면 "완료"라고 말하지 말 것. - knot·요금가드 안내(선택) — 사용자가 knot 지식 vault나 goal 요금가드를 찾으면 알린다: 두 구성의 설치는 v3.0.0부터 loadout 카탈로그(https://github.com/netwaif/loadout) 담당이다("CLAUDE.md 구성 골라 담아줘").
knot능동 스킬(save/ingest/query/lint)은 v3.1.0부터 knot 자체 플러그인이 배포한다(마켓플레이스에netwaif/knot추가). 가드 배선 자산(claude Stop 훅·codex 워처)도 v3.2.0부터 loadout guard 품목이 전부 제공한다(codex는--flavor codex).
동작 보장
- 결정적: 번들 템플릿을 그대로 복사. 모델이 시스템 파일을 창작하지 않는다.
- 안전: 대상에 기존
tasks/·_local/사용자 데이터가 있으면 보존(update 모드). 기존 지침파일(CLAUDE.md/AGENTS.md)은 덮어쓰기 전.multiagent-bak으로 백업하고, loadout store 조각 블록(<!-- store:*:start/end -->)은 새 지침파일 끝에 재부착한다. - 쓰기 권한: 파일 생성이므로 쓰기 권한이 필요하다. Codex에서는
workspace-write+ 승인이 필요할 수 있다 — 막히면 사용자에게 권한을 안내한다.
Do NOT
- 시스템 파일(CLAUDE.md/AGENTS.md,
_shared/*,_templates/*)을 직접 작성·수정하지 말 것. 항상init.py로 생성. - 플러그인 자신의 폴더나
generator/templates/(이 스킬 폴더 안) 안에 설치하지 말 것 (init.py가 막지만 시도도 금지). - validate FAIL을 숨기거나 "대충 됐다"고 보고하지 말 것.
What ships with it
60 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- generator/build_zip.py 8.2 KB runs code
- generator/init.py 7.7 KB runs code
- generator/knot_block.md 453 B
- generator/knot-vault/.gitignore 28 B
- generator/knot-vault/AGENTS.md 196 B
- generator/knot-vault/CLAUDE.md 196 B
- generator/knot-vault/GEMINI.md 196 B
- generator/knot-vault/inbox/.gitkeep 0 B
- generator/knot-vault/index.md 188 B
- generator/knot-vault/log.md 116 B
- generator/knot-vault/prompts/ingest.md 1.6 KB
- generator/knot-vault/prompts/lint.md 1.2 KB
- generator/knot-vault/prompts/query.md 883 B
- generator/knot-vault/raw/.gitkeep 0 B
- generator/knot-vault/README.md 1.6 KB
- generator/knot-vault/schema.md 4.2 KB
- generator/knot-vault/scripts/drain.sh 6.2 KB runs code
- generator/knot-vault/scripts/launchd.plist.template 1.4 KB
- generator/knot-vault/scripts/lint.py 5.8 KB runs code
- generator/knot-vault/wiki/.gitkeep 0 B
- generator/sync_claude_template.py 7.3 KB runs code
- generator/templates/antigravity/_local/.gitkeep 0 B
- generator/templates/antigravity/_shared/adapters/_run.py 1.4 KB runs code
- generator/templates/antigravity/_shared/adapters/call_worker.sh 8.1 KB runs code
- generator/templates/antigravity/_shared/approval-policy.md 2.1 KB
- generator/templates/antigravity/_shared/backends.json 1.4 KB
- generator/templates/antigravity/_shared/capability-profile.md 1.9 KB
- generator/templates/antigravity/_shared/check-invariants.sh 4.5 KB runs code
- generator/templates/antigravity/_shared/design-basis.md 5.9 KB
- generator/templates/antigravity/_shared/learnings.md 1.1 KB
- generator/templates/antigravity/_shared/orchestrator-rules.md 4.4 KB
- generator/templates/antigravity/_shared/routing.md 7.5 KB
- generator/templates/antigravity/_shared/system-invariants.md 2.3 KB
- generator/templates/antigravity/_templates/context.md 751 B
- generator/templates/antigravity/_templates/log.md 1.1 KB
- generator/templates/antigravity/_templates/task-folder.md 4.1 KB
- generator/templates/antigravity/_templates/task.md 1.3 KB
- generator/templates/antigravity/_templates/worker-brief.md 1.7 KB
- generator/templates/antigravity/_templates/worker-result.md 1011 B
- generator/templates/antigravity/.gitignore 381 B
- generator/templates/antigravity/AGENTS.md 9.3 KB
- generator/templates/antigravity/CHANGELOG.md 4.1 KB
- generator/templates/antigravity/KNOWN_ISSUES.md 2.9 KB
- generator/templates/antigravity/NOTICE 2.5 KB
- generator/templates/antigravity/README.md 3.6 KB
- generator/templates/antigravity/tasks/.gitkeep 0 B
- generator/templates/claude/_local/.gitkeep 0 B
- generator/templates/claude/_shared/adapters/_run.py 1.4 KB runs code
- generator/templates/claude/_shared/adapters/call_worker.sh 8.1 KB runs code
- generator/templates/claude/_shared/adapters/gemini_api.sh 976 B runs code
- generator/templates/claude/_shared/approval-policy.md 2.1 KB
- generator/templates/claude/_shared/backends.json 2.8 KB
- generator/templates/claude/_shared/capability-profile.md 1.9 KB
- generator/templates/claude/_shared/check-invariants.sh 13 KB runs code
- generator/templates/claude/_shared/design-basis.md 13 KB
- generator/templates/claude/_shared/learnings.md 14 KB
- generator/templates/claude/_shared/orchestrator-rules.md 6.5 KB
- generator/templates/claude/_shared/routing.md 14 KB
- generator/templates/claude/_shared/system-invariants.md 5.3 KB
- generator/templates/claude/_templates/context.md 751 B
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 5d ago First seen · 38 lines · 118 tokens per session scan A ccb18664b9ea
configure-multiagent is a skill published in the GitHub repository netwaif/multi-agent-starter (102 stars, last pushed 1mo ago), licensed MIT. It adds 118 tokens to every session and 1,061 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
chronicle
Analyze Copilot session history for standup reports, usage tips, session search, and session reindexing. Use when the user asks for a standup, daily summary, usage tips, workflow recommendations, wants to search or find past sessions by keyword/file/PR, wants to reindex their session store, or asks about deleting…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…