AI agents invoke jobhound_pause to trigger actions in JobHound. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.
This tool executes a control operation on the JobHound daemon, stopping its scan/apply loop. It doesn't read, write, or delete data, nor does it move money — it triggers an external system state change (process pause). Severity is low since pausing is reversible and the blast radius of misuse is minimal (jobs stop being applied to temporarily).
From the tool's definition 'Pause the JobHound daemon (scan loop stops applying)' — triggers an external operation that halts a running daemon process
Attacks that exploit this kind of access
Pause the JobHound daemon (scan loop stops applying). It is categorised as a Execute tool in the JobHound MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the JobHound MCP server in PolicyLayer and add a rule for jobhound_pause: 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 JobHound. Nothing to install.
jobhound_pause is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the jobhound_pause 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 jobhound_pause. 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.
jobhound_pause is provided by the JobHound MCP server (null-phnix/jobhound). 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 →