Having this library helps me introduce resilience to my project, but I don't want expected or unexpected failures to occur to test it out

• The library could be wrongly implemented, testing the scenarios is not straight forward
• Mocking failure of a dependency for example a cloud SaaS / PaaS service is not always provided straight away.

Describe your proposed or preferred solution:

1. A way to mock failures of dependencies (any service dependency for example).
2. This should be strict framework that takes care of when to fail based on some external factors - maybe global configuration or some rule.
3. Way to revert easily controlling the blast radius
4. Production grade to run this in production system with automation.

Describe any alternative options you've considered:
Doing it myself, manually disabling the service which supports that and checking if it really breaks the circuit and moves to fallback.

Any additional info?
We can definitely take learning from Netflix Simian Army.