Tokentoll

v0.2.0 Feature

This release adds 2 notable features for engineering teams evaluating rollout.

Published 1mo Model Serving & MLOps

View tool

✓ No known CVEs patched

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

✓ No known CVEs patched in this version

Topics

anthropic cost-optimization devtools github-action llm mlops

+3 more

openai python static-analysis

Summary

AI summary

Updates Resolution patterns now supported, Stats, and val across a mixed release.

Full changelog

What's New

Multi-pass constant propagation engine -- tokentoll now follows variable assignments, class attributes, constructor arguments, dict contents, and **kwargs unpacking to resolve model names and parameters that aren't string literals.

Resolution patterns now supported

Variable assignments: MODEL = "gpt-4o" then create(model=MODEL)
os.getenv() / os.environ.get() fallbacks
Function default parameters
Class attribute defaults: class Config: model = "gpt-4o"
Constructor argument propagation: obj = Cls(val) -> self.attr = param
Dict literal and subscript tracking: d = {"model": "gpt-4o"} or d["model"] = "gpt-4o"
**kwargs unpacking: create(**kwargs) resolves from tracked dict contents

Stats

47 tests, all passing
Zero runtime dependencies
5 SDK detectors: OpenAI, Anthropic, Google GenAI, LiteLLM, LangChain

Install

pip install tokentoll==0.2.0

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 Tokentoll

Get notified when new releases ship.

About Tokentoll

All releases →