Back to evaluations
Draft evaluation
Industrial Operations Capture Analysis with Mastra AI and Claude 4 Opus — evaluation
Evaluates Mastra AI agent's ability to clean corrupted historian logs and fix capture bottlenecks.
Evaluation type
task based
Challenge
Industrial Operations Capture Analysis with Mastra AI and Claude 4 Opus
Difficulty
Intermediate
Rigor
Not declared
The author has not specified a rigor level.
Evaluation overview
How the linked challenge is judged: tasks, benchmarks, and criteria count.
Tasks
1
Benchmarks
0
Criteria
0
Task templates
Inputs and expected outputs.
Task 1
capture_fault_eval
Receives malformed batch logs and outputs clean payloads alongside root cause.
Input format
JSON array of raw records with timestamp, source_id, raw_string
Output format
JSON object with sanitized_records (array), errors_rectified (int), cause (string)