Metadata-Version: 2.4
Name: thaghr
Version: 0.0.1
Summary: Chaos engineering for AI agents. Find the breach before it finds you.
Project-URL: Homepage, https://github.com/thaghr/thaghr
Project-URL: Source, https://github.com/thaghr/thaghr
Author: Taoheed A. Abdulraheem
License-Expression: Apache-2.0
License-File: LICENSE
Keywords: agents,ai,chaos-engineering,fault-injection,llm,reliability,sre
Classifier: Development Status :: 1 - Planning
Classifier: Intended Audience :: Developers
Classifier: Programming Language :: Python :: 3
Classifier: Topic :: Software Development :: Testing
Classifier: Topic :: System :: Monitoring
Requires-Python: >=3.10
Description-Content-Type: text/markdown

# thaghr

**Chaos engineering for AI agents. Find the breach before it finds you.**

ثَغْر (thaghr) is a fortified frontier outpost. ثُغْرَة (thughra) is the breach it
exists to guard. Same root.

thaghr breaks your AI agent on purpose, then tells you whether it survived.
It injects the failures agents actually hit in production, rate limit storms,
corrupted tool results, truncated streams, silent model swaps, runs the agent
many times under those conditions, and reports whether it degrades gracefully
or falls over.

This is a placeholder release reserving the name. The working tool lands at
`0.1.0`.

- Source: https://github.com/thaghr/thaghr
- Licence: Apache-2.0
