Add JSONL dataset format support#131
Merged
christso merged 9 commits intomainEntityProcess/agentv:mainfrom Jan 20, 2026
feat/add-jsonl-dataset-formatEntityProcess/agentv:feat/add-jsonl-dataset-formatCopy head branch name to clipboard
Merged
Add JSONL dataset format support#131christso merged 9 commits intomainEntityProcess/agentv:mainfrom feat/add-jsonl-dataset-formatEntityProcess/agentv:feat/add-jsonl-dataset-formatCopy head branch name to clipboard
christso merged 9 commits intomainEntityProcess/agentv:mainfrom
feat/add-jsonl-dataset-formatEntityProcess/agentv:feat/add-jsonl-dataset-formatCopy head branch name to clipboard
Conversation
c677ea3 to
504a7dc
Compare
Add support for JSONL (JSON Lines) format as an alternative to YAML for evaluation datasets, following industry standards from DeepEval, LangWatch, Hugging Face, and OpenAI. Key features: - Pure JSONL files (one eval case per line) - Optional sidecar YAML for metadata and defaults - Per-case overrides for execution and evaluators - Same file reference resolution as YAML - Fully backward compatible with existing YAML files Benefits: - Memory efficient for large datasets - Git-friendly line-based diffs - Easy programmatic generation and appending - Compatible with standard JSONL tools Includes complete proposal, design doc, implementation tasks, and spec with 8 requirements and 27 scenarios.
Add JSONL (JSON Lines) format as an alternative to YAML for evaluation datasets, following industry standards from DeepEval, LangWatch, and Hugging Face. Key features: - Pure JSONL data format (one eval case per line) - Optional sidecar YAML metadata file for dataset defaults - Per-case overrides for execution, evaluators, and rubrics - Line-by-line parsing with clear error messages - Same validation and file reference resolution as YAML - Full backward compatibility with existing YAML files Benefits: - Streaming-friendly for large datasets - Git-friendly line-based diffs - Easy programmatic generation - Standard tool compatibility (jq, grep, etc.)
- Replace jsonl-format example with basic-jsonl (mirrors basic example) - Add file reference examples in JSONL format - Update eval-builder skill with JSONL format documentation
The eval command's path resolver was only accepting .yaml files, preventing users from running JSONL datasets directly. Updated regex patterns to accept both .yaml/.yml and .jsonl file extensions, and improved error message to mention JSONL support.
e2b774e to
8c9e50d
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Add support for JSONL (JSON Lines) format as an alternative to YAML for evaluation datasets.
Why
Enables large-scale evaluation workflows following industry standards (DeepEval, LangWatch, Hugging Face, OpenAI):
Design