Capture a reusable learning/pattern to learnings.json. Called after key decisions, verdicts, and pipeline milestones.
AI agents use capture_learning to create or update resources in Launch Engine — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Launch Engine environment.
An AI agent can call capture_learning faster than any human can review — one bad instruction and it creates or modifies resources in Launch Engine by the hundred, each call as confident as the last.
Attacks that exploit this kind of access
Capture a reusable learning/pattern to learnings.json. Called after key decisions, verdicts, and pipeline milestones. It is categorised as a Write tool in the Launch Engine MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Launch Engine MCP server in PolicyLayer and add a rule for capture_learning: 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 Launch Engine. Nothing to install.
capture_learning 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 capture_learning 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 capture_learning. 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.
capture_learning is provided by the Launch Engine MCP server (zionhopkins/asset-factory-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.