Install
$ agentstack add skill-insightflo-claude-impl-tools-team-orchestrate ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Unified Orchestration Engine
> Goal: Execute TASKS.md at any scale with the right coordination strategy.
Mode Selection
| Mode | Task count | Infrastructure | Command | |------|-----------|----------------|---------| | auto | 1-50 | Task tool only | /team-orchestrate --mode=auto | | autonomous | any | Task tool + auto-cycle | /team-orchestrate --mode=autonomous | | team | 10-80 | Agent Teams API | /team-orchestrate --mode=team | | thin | 50-200 | Task tool + background | /team-orchestrate --mode=thin |
Auto-selection (when no --mode specified):
IF tasks = 50 → thin
For mode-specific details:
- auto:
references/auto-mode.md— Direct subagent dispatch, worktree-based phases - thin:
references/thin-mode.md— Ultra-minimal context, 76% token reduction - team: Below (this document) — Agent Teams API, 3-level hierarchy
Mode: team (Agent Teams API)
Mandatory Tools (순서대로 호출)
| 순서 | 도구 | 용도 | 필수 파라미터 | |------|------|------|-------------| | 1 | TeamCreate | 팀 생성 + 공유 작업 리스트 | team_name, description | | 2 | TaskCreate | 작업 등록 | subject, description | | 3 | TaskUpdate | 의존성/소유자 설정 | addBlockedBy, owner | | 4 | Agent | 팀메이트 생성 | team_name, name, prompt | | 5 | SendMessage | 에이전트 간 통신 | to, message |
> CRITICAL: Agent를 호출할 때 반드시 team_name을 포함해야 함. 없으면 일반 서브에이전트가 생성되어 통신 불가.
Prerequisite Checks (실행 전 자동 확인)
| # | 항목 | 실패 시 동작 | |---|------|------------| | 1 | TASKS.md 존재 | STOP - "/tasks-init로 생성 먼저" | | 2 | TASKS.md 형식 (deps:, domain:) | STOP - "/tasks-migrate로 변환" | | 3 | Agent Teams 활성화 | AUTO-FIX - install.sh --local --mode=team | | 4 | 프로젝트 훅 설치 | AUTO-FIX - 동일 | | 5 | settings.json 등록 | AUTO-FIX - 동일 | | 6 | governance 문서 존재 | AskUserQuestion - 사용자 선택 |
Check 6 상세:
ls management/project-plan.md management/decisions/ADR-*.md 2>/dev/null | wc -l
≥ 2→ PASSGSD의/gsd:autonomous` 패턴. 모든 incomplete 작업을 자동으로 순회.
실행 흐름
1. TASKS.md 분석 → incomplete 작업 목록
2. 각 작업에 대해:
a. /discuss (필요시) → CONTEXT.md
b. 계획 수립
c. 실행 (deviation rules 적용)
d. 검증 (goal-backward)
3. 모든 작업 완료 또는 블로커 발생까지 반복
사용자 개입 포인트
- Gray area 결정 필요 (AskUserQuestion)
- 검증 실패 시 확인
- 블로커 발생 시 알림
진행 상황 표시
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
AUTONOMOUS MODE
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Progress: 3/10 tasks
Current: T1.3: Implement auth API
Status: executing...
[██████░░░░░░░░░░░░░░] 30%
호출
/team-orchestrate --mode=autonomous
/team-orchestrate --mode=autonomous --from T2.1 # 특정 작업부터
Fallback
- 5분 무응답 → 알림 + 대기
- 3회 연속 실패 → 중단 + 상태 저장
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: insightflo
- Source: insightflo/claude-impl-tools
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.