Fill a sensitive vault field into a browser element via CSS selector. The secret never appears in the agent context.
AI agents use blind_fill to create or update resources in Shellgate — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Shellgate environment.
This tool writes/injects a sensitive credential from a vault into a browser DOM element. It modifies browser state (filling a form field) without executing code or commands, making it a Write action. The severity is medium because it handles sensitive credentials and interacts with browser elements, but the action itself is a controlled, targeted field fill rather than broad execution or destruction.
From the tool's definition Fill a sensitive vault field into a browser element via CSS selector
Documented attack patterns abuse exactly the kind of access blind_fill gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Shellgate, and nothing reaches the server without passing your rules. This is the rule we recommend for blind_fill:
{
"version": "1",
"default": "deny",
"tools": {
"blind_fill": {
"limits": [
{
"counter": "blind_fill_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}blind_fill stays usable, but capped — an agent stuck in a loop can't make hundreds of changes a minute. Everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Fill a sensitive vault field into a browser element via CSS selector. The secret never appears in the agent context. It is categorised as a Write tool in the Shellgate MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Shellgate MCP server in PolicyLayer and add a rule for blind_fill: 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 Shellgate. Nothing to install.
blind_fill is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the blind_fill 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 blind_fill. 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.
blind_fill is provided by the Shellgate MCP server (matthiastjong/shellgate). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Shellgate, 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.
31 Shellgate tools catalogued and risk-classified — across an index of 43,000+ MCP servers.