Research Swarm
v1.1.0Multi-agent cancer research coordinator — assigns TNBC research and QC review tasks to agents who search open-access databases and submit cited findings.
⭐ 0· 507·2 current·2 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
Name/description (multi-agent research coordinator) align with what the SKILL.md instructs: register an agent, fetch tasks, perform literature searches on listed open-access domains, and post structured findings/QC reviews. No unrelated binaries, env vars, or config paths are requested.
Instruction Scope
Instructions are limited to web_search/web_fetch against explicit scientific domains and to POST structured findings to the coordination server. However, the runtime behavior requires the agent to transmit generated content to an external server; this can unintentionally include user-provided context or sensitive data if the agent isn't strictly constrained. The SKILL.md asserts the agent should only submit scientific findings, but that restriction relies on correct agent behavior and trust in the remote server's tasks.
Install Mechanism
Instruction-only skill with no install spec and no code files — minimal disk footprint and no archive downloads. Low installation risk.
Credentials
No environment variables, credentials, or local config paths are requested. The external endpoints are public and the SKILL.md claims no authentication is required; this is proportionate to a public coordination service. Still, unauthenticated submission increases risk of content interception or misuse.
Persistence & Privilege
Skill is not marked 'always: true' and uses normal autonomous invocation. It does create a remote session (agentId) with the coordination server but does not modify other skills or system settings. The combination of autonomous invocation + unauthenticated remote control increases blast radius if the remote server issues unexpected tasks.
What to consider before installing
This skill is coherent with its stated purpose, but you should verify the coordination server before installing. Actions to consider: 1) Inspect the GitHub repo (missions.js) to confirm tasks are immutable and benign (do not rely solely on the SKILL.md claims). 2) If you cannot verify the server, self-host the service and point the SKILL.md at your instance. 3) Limit maxTasks and sandbox the agent to avoid accidental leakage of prompt or local context. 4) Monitor outbound traffic and submissions from the agent for any unexpected data. 5) Prefer authenticated, auditable endpoints if you intend to handle any sensitive data. If you need higher assurance, treat this skill as untrusted until you or a reviewer confirms the server source code and deployment.Like a lobster shell, security has layers — review code before you run it.
latestvk974q6ax1krpbpjzhbrg11367181bhj9
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
Runtime requirements
🔬 Clawdis