Record an outcome or goal success signal for a run. Requires --enable-registry-writes because it writes telemetry.
AI agents invoke record_outcome to trigger actions in Tuning Engines - LLM Fine-Tuning. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.
record_outcome triggers real processes with real consequences. An agent gone sideways doesn't fire it once — it starts dozens of builds, sends mass notifications, or burns through compute before anyone looks up.
Attacks that exploit this kind of access
Record an outcome or goal success signal for a run. Requires --enable-registry-writes because it writes telemetry. It is categorised as a Execute tool in the Tuning Engines - LLM Fine-Tuning MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Tuning Engines - LLM Fine-Tuning MCP server in PolicyLayer and add a rule for record_outcome: 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 Tuning Engines - LLM Fine-Tuning. Nothing to install.
record_outcome is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the record_outcome 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 record_outcome. 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.
record_outcome is provided by the Tuning Engines - LLM Fine-Tuning MCP server (tuningengines-cli). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.