Tools/CSV ➔ Eval

CSV ➔ Eval Dataset

Convert raw query-response CSV files into structured JSONL validation datasets and assertion rubrics in-browser.

DATA_INPUT // csv_reader
EVAL_OUTPUT // jsonl_assertions

Awaiting CSV Data Ingestion...

How It Works

This prototype acts as an offline helper compiling raw query logs. It normalizes comma-separated structures, auto-detects input vs expected headers, and maps them to standard assertion objects suitable for validation loops.

Use Cases

  • Structuring user execution history CSVs into RAG evaluation tables.
  • Generating programmatic validation rubrics from product spreadsheet specs.
  • Bootstrapping assertions for multi-agent validation loops.

Planned CLI

Deploy dataset compilation and assertion rubric tests offline inside your local workspace via our planned command:

Planned install command: pip install tesserakit-core tesserakit-evals
Tool Profile
StatusLocal Prototype
ModeBrowser prototype
EngineToolSmith
ContractTessera-compatible
Future WiringTessera JobPack
OutputJSONL + Rubric draft
Browser Limits

Online uploads are restricted to 10 rows. Local CLI executions run offline with zero restrictions.