I’ve been working on a project called Bashqueues—an opinionated, shell-native approach to interprocess communication (IPC) and job queue management on Linux.
Most existing queueing systems are designed for high-scale distributed tasks, often carrying significant overhead or requiring heavy runtime environments. Bashqueues is built for a different use case: environments where IPC governance, strict security policies, and forensic auditability are the primary requirements.
The core philosophy: Instead of just managing "work," Bashqueues treats every job as an asset that must comply with a defined "Class Policy." We want to ensure that a job running in production is exactly what the operator intended, and nothing more.
Key Features (Current Implementation):
Policy-Driven Governance: Every job is bound to a class definition (e.g., `SECURE_OFFICIAL`, `BATCH_PROCESSING`). Policies dictate sandbox levels (seccomp, namespaces), execution caps, and network egress limits before the job is dispatched.
Static & Runtime Auditing: The system includes
secaudit assets to scan for dangerous patterns, and interrogation profiles to baseline normal system behavior.Shell-Native: The engine (`queuebash.sh`) and management interface (`queuemgr_panel.py`) are designed to be transparent, scriptable, and easy to interrogate using standard POSIX shell tools.
Forensic Readiness: Every dispatch, failure, and policy exception is logged with structured metadata, designed for environments where you need to know exactly why a job was blocked or allowed.
Current State & Disclaimer: This project is currently in active, early-stage development.
Code Stability: It is functional for our internal use cases, but it is not "production-ready" in the sense of enterprise software. Expect to find edge cases, especially regarding complex systemd daemon configurations.
Scope: It is designed for specific, policy-heavy Linux environments. It is not intended to replace high-concurrency message queues (like RabbitMQ or Kafka).
I’m sharing this because I am looking for eyes on the logic—specifically the policy enforcement and security-governance class statements. If you have experience with Linux security hardening, systemd, or shell-based orchestration and want to critique the architecture, I’d appreciate the input.
As the notes make clear, this was designed by a human, but coded by an AI, an AI checked the work, and a variety of other AI's have contributed to this project. So, when someone says "Did ChatGPT write this?" then the answer is yes, Claude checked it, Co-Pilot discussed the Microsoft and other commercial infrastructure, Deepseek gave suggestions and Gemini wrote the majority of the Reddit post.
Repository: https://github.com/animatedads/bashqueues
Note: All feedback regarding security implementation is welcome. Please handle any potential bug reports via the standard GitHub issue tracker.
https://redd.it/1tpfynb
@r_linux
