Novaforge

Diagnose the root cause of a machine anomaly by semantically searching historical maintenance logs. Computes which telemetry readings are actually outside nominal range, then finds the best-matching past incident with a relevance score, the resolution that worked, and cost/downtime data. Use this after identifying a machine with an active anomaly (e.g. from the fleet triage).

Compare the cost and downtime of fixing a diagnosed machine anomaly now versus deferring the fix to the end of the shift. Returns a business-decision comparison with estimated failure risk if deferred, not just a technical recommendation. Call diagnose_anomaly first if you have not already — this tool re-runs the same diagnosis internally.

Execute a physical remediation on a machine (e.g. trigger the actuator once a fix like a coolant filter replacement is confirmed). ALWAYS requires manager approval — calling this without a valid approval code will be refused and a one-time code will be issued. Call again with that code to actually approve and queue the action.

Verify that a previously executed remediation actually resolved the machine anomaly. Re-reads live telemetry, compares it against the pre-remediation snapshot, confirms all readings are back within nominal range, and writes a permanent audit record. Call this after execute_remediation has completed (status: executed).

Generate a downloadable shift handover report for a diagnosed-and-verified incident on a machine. Use this after verify_remediation has confirmed an incident is resolved.