hidai25/eval-view

v0.7.1 Breaking

This release includes 2 breaking changes for platform teams planning a safe upgrade.

Published 1mo Developer Productivity

View tool

✓ No known CVEs patched

Read the diff → Tool health → What is this tool? →

✓ No known CVEs patched in this version

Topics

agent-benchmark agent-evaluation agentic-ai ai-agents anthropic autogen

+12 more

cli crewai evaluation langchain-agent langgraph llm mcp openai-assistants pytest python regression-testing testing

Summary

AI summary

Added TOML test case support and CSV log import to evalview.

Full changelog

What's new in 0.7.1

Added

TOML test cases — TestCaseLoader now understands .toml alongside .yaml/.yml (#209). Same schema, different on-disk encoding for teams that already author config in TOML. Uses stdlib tomllib on Python 3.11+ and falls back to tomli on 3.9/3.10. Example in examples/test_case_toml.toml.
CSV log import — evalview generate --from-log now accepts CSV alongside JSONL, OpenAI, and EvalView capture formats (#216). Header row identifies columns, with aliases matching the JSONL parser. Tool cells accept JSON list, comma-, semicolon-, or pipe-separated forms. Example fixture and README under examples/log-import/.

Changed

Chat-mode command validator now derives its allowlist from the Click registry instead of a hand-maintained set.
evalview inspect is now a hidden alias that forwards to evalview visualize (which is the strict superset).
evalview view is no longer hidden — surfaced under the "Inspect & Visualize" section of --help.
Daily dogfood workflow now updates a single rolling issue instead of filing a new one each failing day.
parse_csv's warn callback type tightened from Any to Optional[Callable[[str], None]] (#221).

Removed

evalview quickstart deprecated shim (use evalview demo or evalview init).
requirements.txt (use pyproject.toml + uv.lock).

Internal

Four 1k+ files split into focused submodules following the PR #202 mixin pattern (#215).
Root-directory polish pass (#218): tighter .gitignore, install.sh → scripts/install.sh, 39 tracked root entries.

Docs

AGENT_INSTRUCTIONS.md → AGENTS.md, and guides/ moved under docs/ (#219).
docs/README.md index expanded; per-directory READMEs in demo/, demo-agent/, demo-tests/.

Fixed

Hermetic test suite — git clone && make test goes green on first run with no env setup beyond uv.
Lint-clean main — drops unused imports left after the quickstart removal.

PyPI: https://pypi.org/project/evalview/0.7.1/
npm: https://www.npmjs.com/package/evalview/v/0.7.1
Full changelog: https://github.com/hidai25/eval-view/blob/main/CHANGELOG.md

🤖 Co-authored with Claude Code

Breaking Changes

Removed `evalview quickstart` shim (use `evalview demo` or `evalview init`).
Removed `requirements.txt` (use `pyproject.toml` + `uv.lock`).

View diff on GitHub

Weekly OSS security release digest.

The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.

No spam, unsubscribe anytime.

Share this release

Share on X Share on Bluesky

Track hidai25/eval-view

Get notified when new releases ship.

About hidai25/eval-view

Regression testing framework for AI agents. Save golden baselines, detect behavioral drift, and block regressions in CI. Works with LangGraph, CrewAI, OpenAI, Claude, and any HTTP API.

All releases →