Agent Swarm Planner
v1.0.0Designs complete multi-agent system architectures specifying agent roles, communication, memory, escalation protocols, schedules, and risk maps for reliable...
⭐ 0· 35·0 current·0 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 and description (multi-agent architecture design) match the actual contents: templates, design principles, failure patterns, and a runnable design workflow. No unrelated binaries, env vars, or config paths are requested.
Instruction Scope
SKILL.md contains prescriptive guidance for designing swarms and templates; it does not instruct the agent to read local files, access environment variables, call external endpoints, or transmit user data. All steps are confined to producing architecture documents and asking clarifying questions.
Install Mechanism
No install spec or code files to write/execute. Being instruction-only means nothing is downloaded or installed on the host—low persistence and low supply-chain risk.
Credentials
No credentials, secrets, or config paths are requested. The templates reference external channels (Slack/email) only as theoretical escalation targets within the architecture, not as required variables or endpoints for the skill itself.
Persistence & Privilege
always is false and the skill is user-invocable; model invocation is allowed (the platform default). The skill does not request elevated or permanent agent presence or modify other skills/configuration.
Assessment
This skill appears coherent and safe as a planning/template tool. Before installing, confirm you trust the publisher (source/homepage is not provided), and treat outputs as design guidance—not as code to run automatically. When you implement any swarm described by this skill, enforce strict human-in-the-loop controls, limit agent credentials/permissions per the templates' least-privilege advice, test in a sandbox, and require explicit human approval before any agent performs real-world actions or is given access to production systems.Like a lobster shell, security has layers — review code before you run it.
latestvk976c9rp7mp104d5w55jwjzz7h842av0
License
MIT-0
Free to use, modify, and redistribute. No attribution required.