AI agents use upsert_partner to create or update resources in Cockpit — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Cockpit environment.
The tool creates or updates partner records in a financial accounting system (freee). While reversible (can be corrected via corrections or deletions in typical accounting software), upsert operations that auto-create master data records in accounting contexts pose significant operational risk if misused by an AI agent (e.g., creating duplicate or fraudulent partner entries that inflate transaction records or…
From the tool's definition Tool performs 'auto-creation in freee' and 'create' operation with 'upsert' (update or insert) semantics on partner master data (取引先マスタ). This modifies accounting data reversibly.
Attacks that exploit this kind of access
取引先マスタ auto-creation in freee. Fuzzy match against existing partners; if new, create. Returns partner_id. It is categorised as a Write tool in the Cockpit MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Cockpit MCP server in PolicyLayer and add a rule for upsert_partner: 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 Cockpit. Nothing to install.
upsert_partner 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 upsert_partner 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 upsert_partner. 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.
upsert_partner is provided by the Cockpit MCP server (michielinksee/kansei-link-cockpit). 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 →