← Back to Skills Marketplace
teamarchitect-skill
by
MarsHong-86
· GitHub ↗
· v1.0.0
· MIT-0
124
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install teamarchitect-skill
Description
通过对话收集需求，设计多智能体团队架构，并推荐对应 Agent 角色的专业技能供选择和安装。
Usage Guidance
This skill appears to do what it says (design teams and recommend/install skills), but it needs shell and network access and will read/write your skills directory to fetch or create skills. Before installing: 1) Prefer not to grant autonomous-installation privileges — require manual confirmation for every install or disable autonomous invocation for this skill. 2) Inspect any recommended skill's source (ClawdHub/GitHub) before approving install; check maintainers and code. 3) Restrict its filesystem scope if possible (limit to a dedicated skills folder). 4) Note the README suggests cloning from a GitHub repo (MarsHong-86) — verify that URL and the repo's contents. 5) If you cannot audit downstream skill code, treat installs as higher risk and avoid enabling automatic batch installs.
Capability Analysis
Type: OpenClaw Skill
Name: teamarchitect-skill
Version: 1.0.0
The TeamArchitect skill bundle is a design and orchestration tool for multi-agent systems. It provides a structured workflow for requirement gathering, team architecture design, and skill recommendation across various roles (e.g., analysis, execution, review). The instructions in SKILL.md and the mapping in skills-index.md are entirely aligned with the stated purpose of helping users organize and install relevant skills. No malicious code, data exfiltration, or harmful prompt injection attempts were identified.
Capability Assessment
Purpose & Capability
The skill's purpose (design multi-agent team and recommend/install skills) aligns with its behavior: it checks local skills, queries ClawdHub/GitHub, and may create new skills. The meta.json permissions (bash, network, filesystem: skills-directory) are consistent with installation duties. Minor inconsistency: registry summary lists no required config paths, but SKILL.md and README reference ~/.claude/skills/ and meta.json grants filesystem access.
Instruction Scope
SKILL.md stays within stated purpose (collect requirements, map roles to skills, check local installations, fetch from ClawdHub/GitHub, create new skills). It instructs reading the user's skills directory and fetching from external sources — which is expected for an installer but expands the agent's scope to downloading and writing code. Instructions do not specify validation or allowlisting of external sources.
Install Mechanism
There is no packaged install spec (instruction-only), but README and SKILL.md explicitly instruct using network actions (git clone, ClawdHub downloads) and meta.json allows bash/network. Downloading and writing external skill code is plausible and not automatically vetted; the install mechanism therefore carries supply‑chain risk because arbitrary repositories or user‑specified sources may be executed/installed.
Credentials
The skill does not request environment variables or secret credentials. That is proportionate to its stated function. However, the skill requests network and shell permissions (via meta.json) which are powerful even without secrets because they allow fetching and executing external code.
Persistence & Privilege
always:false (good). But the skill is allowed to read/write the skills directory and run shell/network actions and the platform default permits autonomous invocation. Combined, these privileges could let the agent autonomously download and install additional skills if given consent or if it acts autonomously — increasing the blast radius if a malicious repo is used.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install teamarchitect-skill - After installation, invoke the skill by name or use
/teamarchitect-skill - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
TeamArchitect 1.0.0 initial release:
- Enables multi-turn, natural language conversations to gather project requirements and design multi-agent team architecture.
- Dynamically recommends essential and optional skills for each Agent role, based on user input.
- Provides interactive installation guides for required and enhanced skills, with options for local, marketplace, open-source, or on-demand creation.
- Includes built-in logic for skill compatibility checks and configuration optimization.
- Features example dialogues to guide users through team design and skill installation.
Metadata
Frequently Asked Questions
What is teamarchitect-skill?
通过对话收集需求，设计多智能体团队架构，并推荐对应 Agent 角色的专业技能供选择和安装。 It is an AI Agent Skill for Claude Code / OpenClaw, with 124 downloads so far.
How do I install teamarchitect-skill?
Run "/install teamarchitect-skill" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is teamarchitect-skill free?
Yes, teamarchitect-skill is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does teamarchitect-skill support?
teamarchitect-skill is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created teamarchitect-skill?
It is built and maintained by MarsHong-86 (@marshong-86); the current version is v1.0.0.
More Skills