OSS-Fuzz integration does not work for the majority for the projects

I have been trying to identify new interesting benchmarks to evaluate a fuzzer. For this I wrote [some scripts](https://github.com/florian-university/format_fuzzer_experiments) that integrate all OSS-Fuzz experiments to fuzzbench. It basically runs the oss_fuzz_benchmark_integration.py and tests that integration by running `make test-run-afl-{benchmark_name}`.

After integrating all, only about 40 out of >300 projects succeed.

I'm not quite sure, whether this is because of wrong integration on my part or broken experiments.

Thank you very much.