
PoliFlow PoliFlow provides a workflow-aware approach to enforce control-flow policies in serverless applications by extracting allowed execution paths from workflow descriptors and enforcing them at runtime on Knative Services. This repository aggregates and links to the open-source components: Extractor, Enforcer, Language, Test Applications, and Evaluation assets. Extractor The repository contains two Extractor versions: CNCF Serverless Workflow: Parses v0.8 descriptors (and optional subflows) to produce allowed-paths in JSON (for enforcement) and YAML (human-readable); examples and pre-extracted outputs are included. PoliFlow Language: For direct-call applications (function-to-function/services), outputting inbound and outbound rules; requires installing the PoliFlow language package locally before running. The repository is accessible using the URL https://doi.org/10.5281/zenodo.18625795 Enforcer The implemented Enforcer works as an extension to Knative's queue-proxy that enforces Extractor-produced allowed paths (for the CNCF Serverless Workflow v0.8) on each protected Knative Service. The repository is accessible using the URL https://doi.org/10.5281/zenodo.18625781 Language The PoliFlow Declarative Language is used for defining control-flow policies in serverless direct-call applications. Moreover, this repository holds a Python package used by the PoliFlow Language Extractor. The repository is accessible using the URL https://doi.org/10.5281/zenodo.18625802 Test Applications This repository implements minimal applications used to validate enforcement across workflow patterns (loop, parallel, operation, conditional) with Knative Services and SonataFlow-based workflows. Each application includes allowed-path annotations extracted from the workflow and can be modified to verify that the Enforcer blocks unintended flows. The repository is accessible using the URL https://doi.org/10.5281/zenodo.18625740 Evaluation The Evaluation repository holds the scripts for deployment/teardown timing and latency measurements across several scenarios (Refund, Valve, long-sequence, long-parallel) used in the PoliFlow article "PoliFlow: Inferring Control-Flow Policies from Serverless Workflows." The repository is accessible using the URL https://doi.org/10.5281/zenodo.18625790
Function as a Service, FaaS, CFI, Control Flow Integrity, Cloud Computing, Serverless Computing, Computer Security
Function as a Service, FaaS, CFI, Control Flow Integrity, Cloud Computing, Serverless Computing, Computer Security
| selected citations These citations are derived from selected sources. This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically). | 0 | |
| popularity This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network. | Average | |
| influence This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically). | Average | |
| impulse This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network. | Average |
