AI agents use add_to_seed to create or update resources in SEED MCP — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your SEED MCP environment.
The tool modifies SEED's internal knowledge/data store by adding to it. This falls under Write (creates/modifies data reversibly) rather than Read (passive retrieval) or more severe categories. While the blast radius is medium due to potential pollution of knowledge systems, it lacks the irreversibility of Destructive operations or the code execution risk of Execute operations.
From the tool's definition Tool name 'add_to_seed' and description 'Instructions for extending SEED's knowledge' indicate creation or modification of SEED's knowledge base. This is a write operation that alters system state but appears reversible.
Attacks that exploit this kind of access
Instructions for extending SEED's knowledge. It is categorised as a Write tool in the SEED MCP MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the SEED MCP server in PolicyLayer and add a rule for add_to_seed: 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 SEED MCP. Nothing to install.
add_to_seed 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 add_to_seed 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 add_to_seed. 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.
add_to_seed is provided by the SEED MCP server (sancovp/seed-mcp). 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 →