Pushes raw HTML content to a display, immediately replacing whatever is currently shown to viewers. Prefer this over send_url unless the user explicitly wants an external page rendered as-is. Include a human-readable description whenever possible so later get_display_content calls can summarize i...
Risk signalsAccepts raw HTML/template content (html) · Handles credentials or secrets (token)
Part of the Agentview server.
Free to start. No card required.
AI agents use send_html to create or modify resources in Agentview. Write operations carry medium risk because an autonomous agent could trigger bulk unintended modifications. Rate limits prevent a single agent session from making hundreds of changes in rapid succession. Argument validation ensures the agent passes expected values.
Without a policy, an AI agent could call send_html repeatedly, creating or modifying resources faster than any human could review. PolicyLayer's rate limiting ensures write operations happen at a controlled pace, and argument validation catches malformed or unexpected inputs before they reach Agentview.
Write tools can modify data. A rate limit prevents runaway bulk operations from AI agents.
{
"version": "1",
"default": "deny",
"tools": {
"send_html": {
"limits": [
{
"counter": "send_html_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}See the full Agentview policy for all 55 tools.
These attack patterns abuse exactly the kind of access send_html gives an agent. Each links to the full case and the policy that stops it:
Other write tools across the catalogue. The same approach applies to each: rate-limit and validate the arguments.
Pushes raw HTML content to a display, immediately replacing whatever is currently shown to viewers. Prefer this over send_url unless the user explicitly wants an external page rendered as-is. Include a human-readable description whenever possible so later get_display_content calls can summarize intent without immediately reading raw HTML. Before generating complex HTML, inspect get_display_capabilities for the target display so you know the real browser/runtime limits. When the user does not supply a detailed design system, default to premium digital-signage quality: full-screen layout, strong visual hierarchy, atmospheric background treatment, refined typography, robust fallback data, and no action buttons unless touch interaction is explicitly requested. For the full creative brief, load render_premium_display_html or read agentview://public/design-system before generating the HTML. Requires authentication with at least content_only scope; the user must be allowed to manage the target display. Exactly one of html or base64_html must be provided. Returns id, name, duration, file (stored filename) and version (content version ID).. It is categorised as a Write tool in the Agentview MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Agentview MCP server in PolicyLayer and add a rule for send_html: 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 Agentview. Nothing to install.
send_html 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 send_html 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 send_html. 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.
send_html is provided by the Agentview MCP server (rafaelkocurek-nvob/agentview). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 55 Agentview tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
4,600+ MCP servers and 31,000+ tools scanned and risk-classified.