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 skills add tellang/triflux --skill tfx-multigit clone --depth 1 https://github.com/tellang/trifluxWrote 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/tellang/triflux/tfx-multi)<a href="https://agentmods.dev/skills/tellang/triflux/tfx-multi"><img src="https://agentmods.dev/badge/skills/tellang/triflux/tfx-multi/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/tellang/triflux/tfx-multi"><img src="https://agentmods.dev/badge/skills/tellang/triflux/tfx-multi.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00035 | $0.06204 |
| Opus 5 | $0.00017 | $0.03102 |
| Sonnet 5 | $0.00007 | $0.01241 |
| Haiku 4.5 | $0.00003 | $0.00620 |
Grade C, and why
tfx-multi scanned grade C with 2 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 8d 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
rm -rf ~/.claude/teams/{teamName}/ ~/.claude/tasks/{teamName}/ Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- `curl -sf http://127.0.0.1:27888/status >/dev/null && test -f ~/.claude/scripts/tfx-route.sh && echo "preflight: ok" || echo "preflight: FAIL"` How it starts
The opening of the file, as written. The whole thing — 427 lines — stays where its author put it; the contents beside it link to each section on GitHub.
tfx-multi v3 — 파이프라인 기반 멀티-CLI 팀 오케스트레이터
Claude Code Native Teams의 Shift+Down 네비게이션을 복원한다. Codex/Antigravity 워커마다 최소 프롬프트(~100 토큰)의 슬림 Agent 래퍼를 spawn하여 네비게이션에 등록하고, 실제 작업은
tfx-route.sh가 수행한다. task 상태는team_task_list를 truth source로 검증한다. v3 —--quick(기본, v2.2 호환) +--thorough(전체 파이프라인: plan→prd→exec→verify→fix loop).
[필수] Lead 고토큰 MCP 직접 사용 금지 Lead(Claude Opus)는 다음 MCP 도구를 직접 호출하지 마라:
- 웹 서치: brave-search, exa, tavily
- 외부 서비스: Notion (notion-), Jira/Confluence (Atlassian mcp__claude_ai_Atlassian__), Google Calendar, Gmail
- 대량 조회: 다중 페이지 읽기, 배치 검색, 이슈 목록 조회 등 반복적 MCP 호출
위임 방법:
- 웹 서치/리서치 →
scientist또는document-specialist역할의 Codex 워커- Notion 대형 페이지 →
bash ~/.claude/scripts/notion-read.mjs --delegate(다운로드 후 로컬 처리)- Jira/Confluence 배치 조회 → Codex
scientist워커에 위임 (워커가 MCP 직접 호출)- 단건 조회(이슈 1개 확인 등) → Lead 직접 허용하되, 3회 이상 반복 호출 시 위임 전환
Lead가 MCP 도구에 직접 접근 가능하더라도, 위임해야 Claude 토큰을 절약할 수 있다. 특히 Notion/Jira는 응답이 크고 반복 호출이 잦아 토큰 소모가 매우 높다.
사용법
/tfx-multi "인증 리팩터링 + UI 개선 + 보안 리뷰" # --quick (기본)
/tfx-multi --thorough "인증 리팩터링 + UI 개선 + 보안 리뷰" # 전체 파이프라인
/tfx-multi --agents codex,gemini "프론트+백엔드"
/tfx-multi --tmux "작업" # 레거시 tmux 모드
/tfx-multi status
/tfx-multi stop
실행 워크플로우
Phase 0: 사전 점검 정책 (출력 최소화 + 즉시 spawn)
[필수] Agent spawn 지연 금지 — preflight와 Agent 생성을 병렬로 실행한다. Lead가 preflight를 순차 완료한 뒤 Agent를 spawn하면 사용자 체감 지연이 발생한다. 수동 모드(
N:agent)에서는 입력 파싱 직후 TeamCreate + Agent spawn을 먼저 시작하고, preflight는 동시에 병렬로 수행한다. preflight 실패 시에만 Agent를 중단한다.
- 원칙:
- 수동 모드: Phase 1 파싱 → Phase 3a~3c(TeamCreate + Agent spawn) + Phase 0(preflight) 동시 병렬
- 자동 모드: Phase 0(preflight) + Phase 2(triage) 동시 병렬 → Phase 3
- 사용자가 요청하지 않으면
Searched for ..., 개별Bash(...)로그를 전면 보고하지 않는다. - 리드에는 요약 한 줄만 노출한다. 예:
preflight: ok (route/hub) - Hub 점검은
/status를 기준으로 한다 (/health단독 판정 금지).
- 권장 체크 예시 (단일 Bash로 통합):
curl -sf http://127.0.0.1:27888/status >/dev/null && test -f ~/.claude/scripts/tfx-route.sh && echo "preflight: ok" || echo "preflight: FAIL"
- 실패 시에만 상세를 노출한다:
tfx-route.sh없음- Hub 비정상/미기동
- 필수 CLI 미설치
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.
- 8d ago First seen · 427 lines · 35 tokens per session scan C 3966938debab
tfx-multi is a skill published in the GitHub repository tellang/triflux (7 stars, last pushed 6d ago), licensed MIT. It adds 35 tokens to every session and 6,204 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it C with 2 findings (recursive force delete, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
ready
Make an unfamiliar repository agent-ready with a verify command, smoke tests, guardrails, and project notes. Use for repository setup, agentic engineering, verify commands, hooks, or allowlists; skip when verification and scope are already clear.
handoff
Write a state capsule so the next session — or a different agent/harness — can resume unfinished work without re-deriving context. Use when a session is ending with work incomplete, when the user says "handoff", "pack up", "ส่งต่อ", "continue tomorrow", when switching between Claude Code and Codex mid-task, or when…
lucia-chat
Set two agent sessions talking through the Luciazero Agent Bus and watch it live in a terminal: pick the pair, open the panes, read the transcript. Use for "ให้ codex กับ claude คุยกัน" or "watch the bus".
done
Run the closeout ritual before handing back non-trivial work; full verification, revert-probe honesty, independent review, and scope reporting. Use before declaring completion, opening a PR, wrapping up a change, or "ปิดงาน".
retro
Record durable lessons, null results, and footguns after hard work or debugging. Use when the user asks for a retro, dead ends need preserving, a task disproves an approach, or "จดบทเรียน". Keep repo knowledge separate from machine-local memory.
show
Visualize code structure, changes, and verification evidence in the smallest useful view. Use for connections, flows, diffs, file maps, Mermaid diagrams, evidence maps, or focused HTML; show facts and label unknowns.