AI agents use sdd_save to create or update resources in Forgespec — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Forgespec environment.
This tool writes data to a spec/contract system and updates phase tracking. While it modifies state, the action is reversible (contracts can be updated again in subsequent phases) and does not delete, destroy, or execute arbitrary code. It falls under Write rather than Execute because it saves validated data rather than running external operations.
From the tool's definition The tool description states 'persist an SDD contract' and 'records the phase transition', indicating it creates or modifies data (the contract and phase state) in a structured, reversible manner.
Attacks that exploit this kind of access
Validate and persist an SDD contract. Records the phase transition for project traceability. It is categorised as a Write tool in the Forgespec MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Forgespec MCP server in PolicyLayer and add a rule for sdd_save: 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 Forgespec. Nothing to install.
sdd_save 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 sdd_save 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 sdd_save. 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.
sdd_save is provided by the Forgespec MCP server (lleontor705/forgespec-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 →