AI agents call job_search to retrieve information from Ashby MCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The job_search tool retrieves or queries job data based on title or criteria without side effects. It matches the Read category pattern of search/list/get operations. Severity is low because job search results are informational only and expose no sensitive vulnerability even if misused by an AI agent.
From the tool's definition Tool description explicitly states 'Search for jobs' with no mention of modification, deletion, or execution. Sibling tools on the same server use clear action verbs (create, update, change_stage, add_tag) that are absent here.
Attacks that exploit this kind of access
Search for jobs by title or other criteria. It is categorised as a Read tool in the Ashby MCP MCP Server, which means it retrieves data without modifying state.
Register the Ashby MCP server in PolicyLayer and add a rule for job_search: 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 Ashby MCP. Nothing to install.
job_search 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 job_search 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 job_search. 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.
job_search is provided by the Ashby MCP server (jaketeagle/ashby-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →