Multi Agent Roles
PassAudited by ClawScan on May 1, 2026.
Overview
This is an instruction-only role-template skill with no code, install steps, credentials, or hidden execution behavior shown.
This skill appears safe to install as an instruction-only template library. If you use it to build real multi-agent workflows, especially in healthcare, finance, or other sensitive domains, add human review, data-sharing rules, and clear limits on what each agent may do.
Findings (2)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
A user could over-rely on the role template for medical decision workflows if they do not add qualified human review and compliance controls.
The healthcare template describes diagnostic and treatment recommendation roles, which are high-trust use cases if someone implements them directly.
"description": "Analyzes patient data and provides diagnostic recommendations" ... "Provide preliminary diagnoses", "Recommend treatment options"
Use these templates only as generic design examples; add explicit human oversight, domain expert validation, and legal/compliance safeguards for healthcare or other regulated uses.
If users turn the templates into a real multi-agent system without adding boundaries, agents may share more information than intended.
The templates define inter-agent communication relationships but do not specify identity checks, data boundaries, or permissions; this is expected for a role-design template but matters if implemented.
"communication": { "inbound": ["role1", "role2"], "outbound": ["role3", "role4"] }When implementing these templates, define which agents may exchange which data, require user approval for sensitive transfers, and separate regulated or private information by role.