Team Collab
v3.1.0多模型团队协作系统 - 像公司一样运作，支持三个专业团队：软件开发团队（8人）、搞钱特战队（10人）、小说+漫剧团队（10人）。固定子窗口模式，每个角色分配最优模型。触发词："软件开发团队"、"搞钱特战队"、"小说团队"。
⭐ 0· 113·1 current·1 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description (multi-model team collaboration) match the SKILL.md, config, and scripts. The package.json, config/default.json, references, and scripts all implement role/model assignment, spawn-call examples, templates and project scaffolding consistent with the stated purpose. No unrelated services or credentials are requested.
Instruction Scope
SKILL.md instructs organizing roles, using triggers, and spawning model-specific sub-windows; the included scripts only print spawn instructions or create local project scaffolding. There are no instructions to read arbitrary system files, exfiltrate data, contact unknown endpoints, or access environment variables beyond what the platform would normally provide.
Install Mechanism
No install spec is provided (instruction-only style with a couple of utility scripts). The files are static and the scripts perform local I/O only (directory creation and printing). No downloads from external URLs or archive extraction are present.
Credentials
The skill declares no required env vars or credentials, which is consistent with the code (no direct network/curl/auth usage). It assumes the hosting platform supplies model/session APIs (the scripts show sessions_spawn usage as examples). If model providers used (e.g., bailian/qwen-*) require separate credentials on your platform, that is an external concern but not an inconsistency in the skill itself.
Persistence & Privilege
always is false and the skill does not request system-level persistence or modify other skills. The README references optional config file locations (~/.openclaw/team-collab.json and ./team-collab.json) which are reasonable for user configuration and not privileged by themselves.
Assessment
This skill appears coherent and implements what it claims: role-based model assignments, spawn-call examples, templates, and simple local project scaffolding. Before installing: 1) Confirm your OpenClaw/platform account already provides access to the model IDs referenced (bailian/qwen-*, deepseek-coder); if those models require provider credentials, ensure you configure them through the platform rather than embedding secrets. 2) Inspect the package repository (package.json points to a GitHub URL) if you plan to run the scripts locally. 3) The included scripts are benign (create folders, write JSON, print spawn examples) — avoid running any code you do not trust in environments that contain sensitive data. 4) If you expect the skill to spawn models autonomously, confirm platform permissions and quotas—autonomous model spawning is normal but increases resource/attack surface. Overall, no hidden network calls, credential access, or obfuscated code was found in these files.Like a lobster shell, security has layers — review code before you run it.
latestvk9770yckg3417yvvm4f0qw6men83kga9
License
MIT-0
Free to use, modify, and redistribute. No attribution required.