AI agents call search_incidents to retrieve information from Rootly without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Search operations are fundamentally retrieval-focused with no side effects. The naming pattern and context of related read-only tools (get, find, collect) strongly suggest this performs incident data queries. Even if misused by an agent, a search cannot modify or delete data, execute arbitrary code, or cause financial harm. Severity is low due to limited blast radius.
From the tool's definition Tool name 'search_incidents' indicates a query/search operation. Sibling tools include 'get_incident', 'find_related_incidents', 'collect_incidents', and 'get_oncall_schedule_summary', which are all read-only operations on incident data.
Documented attack patterns abuse exactly the kind of access search_incidents gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Rootly, and nothing reaches the server without passing your rules. This is the rule we recommend for search_incidents:
{
"version": "1",
"default": "deny",
"tools": {
"search_incidents": {}
}
}search_incidents is read-only, so it stays allowed — but everything else on the server is denied unless you say otherwise.
Free to start. No card required.
search_incidents. It is categorised as a Read tool in the Rootly MCP Server, which means it retrieves data without modifying state.
Register the Rootly MCP server in PolicyLayer and add a rule for search_incidents: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Rootly. Nothing to install.
search_incidents is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the search_incidents rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for search_incidents. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
search_incidents is provided by the Rootly MCP server (https://mcp.rootly.com/sse). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Rootly, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
20 Rootly tools catalogued and risk-classified — across an index of 43,000+ MCP servers.