Security audit
Security checks for vulnerabilities and agentic risk
This skill is a disclosed read-only homes.com lookup helper, with no artifact evidence of hidden commands, persistence, exfiltration, or destructive behavior.
Before installing, be aware that this uses your signed-in homes.com browser session through a separate bridge extension and an npm MCP package. Install only if you are comfortable giving that MCP read access to homes.com pages and saved homes/searches visible in that session, and verify the external package and extension source or checksums first.
Skill accesses MCP server configuration files (mcp.json). MCP configs contain server URLs, authentication tokens, and tool definitions — reading them allows the skill to discover and potentially abuse other tool integrations.
.mcp.json (project) or ~/.claude/mcp.json (global):
{
The manifest says the skill triggers on specific phrases "or any request involving homes.com properties, prices, history, or photos." That fallback condition is broad and ambiguous, making it unclear which requests should or should not invoke the skill beyond the listed examples.
Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.
## Response shape (`view`)
Five tools take `view: "compact" | "full"`, and **`compact` is the default** —
you get the slim shape without asking for it: `homes_search_properties`,
`homes_get_property`, `homes_bulk_get`, `homes_compare_properties`,
`homes_get_market_report`.
No suspicious patterns detected.