AI agents call groundChart to retrieve information from Semiotic without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool constructs a natural-language grounding payload (layered descriptions and communicative-act sentences) about a chart. This is a read/analysis operation that produces descriptive output without creating, modifying, deleting, or executing anything. Severity is low because misuse would at worst produce misleading chart descriptions, not system-level harm.
From the tool's definition 'Build the agent-reader grounding payload' and 'natural-language description' and 'L4 communicative-act sentence' — this tool generates descriptive/explanatory text about a chart without modifying or executing anything
Attacks that exploit this kind of access
Build the agent-reader grounding payload for a Semiotic chart: the layered L1–L3 natural-language description, the L4 communicative-act sentence (what the chart is asking the reader to do —. It is categorised as a Read tool in the Semiotic MCP Server, which means it retrieves data without modifying state.
Register the Semiotic MCP server in PolicyLayer and add a rule for groundChart: 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 Semiotic. Nothing to install.
groundChart 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 groundChart 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 groundChart. 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.
groundChart is provided by the Semiotic MCP server (semiotic). 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 →