AI agents invoke pdfhell_run to trigger actions in Multivon. 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.
Absence of descriptive text reduces confidence, but the tool name and server context indicate this triggers execution of some evaluation or PDF processing operation. Treating as Execute rather than Read because '_run' verbs typically denote active computation or external operations whose side effects depend on arguments (PDF contents, evaluation parameters).
From the tool's definition Tool name 'pdfhell_run' suggests execution of an external operation; context indicates this server provides 'direct access to evaluation tools' for AI agents.
Attacks that exploit this kind of access
pdfhell_run. It is categorised as a Execute tool in the Multivon MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Multivon MCP server in PolicyLayer and add a rule for pdfhell_run: 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 Multivon. Nothing to install.
pdfhell_run 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 pdfhell_run 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 pdfhell_run. 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.
pdfhell_run is provided by the Multivon MCP server (multivon-ai/multivon-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
pdfhell_run is one line of Multivon's registry record.
The record carries the whole server: verified identity, auth posture, risk grade, every tool classified, recommended policy — re-checked continuously.
Teams ship this data inside their own products. See what a licence covers →