AI agents use save_architectural_decision to create or update resources in Handoff — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Handoff environment.
This tool writes/creates new records in a shared memory hub used for project context persistence. While the action is reversible (records can be updated or removed), it modifies shared state that multiple AI agents may depend on for decision-making. Misuse could inject incorrect architectural decisions, misleading future agents about technical choices.
From the tool's definition Tool name contains 'save' and description states 'Save an Architecture Decision Record (ADR)', indicating it creates or modifies persistent project documentation.
Attacks that exploit this kind of access
Save an Architecture Decision Record (ADR). Documents WHY a technical choice was made —. It is categorised as a Write tool in the Handoff MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Handoff MCP server in PolicyLayer and add a rule for save_architectural_decision: 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 Handoff. Nothing to install.
save_architectural_decision 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 save_architectural_decision 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 save_architectural_decision. 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.
save_architectural_decision is provided by the Handoff MCP server (juan-severiano/handoff-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 →