Unverified publisher

Agent Skill & Config Security Audit

io.github.tylerscomic-lab/agent-skill-audit-mcp

Scan AI agent skills and configs for hidden Unicode, prompt injection and exfiltration.

Connection details

MCP serverRemote endpoint listed

Setup & evidence

A remote endpoint is listed. Check publisher instructions for authentication and client support.

Based on registry metadata. Resource type has not been independently reviewed.

Documentation only · not runtime-tested.

Trust & access Review before connecting

Publisher: Publisher not verified
Connection: Remote endpoint listedhttps://agent-skill-audit-mcp.mcpize.run/mcp
Authentication & scopes: Not inspected. Review the server’s authentication requirements before granting access.
Tool permissions: Not inspected. Inspect available tools in a compatible MCP client before use; tool annotations are publisher claims.
Source: MCP Registry ↗ · Repository ↗Catalog observed 2026-10-17io.github.tylerscomic-lab/agent-skill-audit-mcp

Publisher verification is not a security audit. MCP Connect exports configuration; your MCP client and server must enforce permissions and approvals.

https://agent-skill-audit-mcp.mcpize.run/mcp

Follow the publisher’s setup instructions for authentication and client compatibility.

Catalog powered by MCP Connect, by the Apideck team. Sources & credits →