What an OSIRIS JSON producer is expected to do with your infrastructure?

OSIRIS JSON producers are made to showcase that the OSIRIS JSON Specification is not theory but a real thing. An OSIRIS JSON producer is not just a parser like others that happens to emit a bulk raw JSON, it is the boundary where proprietary infrastructure inventories are translated into a portable OSIRIS JSON document that other tools can consume to:

1. Feed or reason with AI platforms and agent under a total control knowing exactly 100% of what you share
2. Create automated accurate reports of your infrastructure in the format of your choice
3. Create high quality draw.io or mermaidjs topologies
4. Creating diff-view comparing two or more OSIRIS JSON documents
5. Identify potential infrastructure drifts
6. Evidence at a point in time for audits or reports
7. Feed your CMDB/IPAM/DCIM with accurate data

The OSIRIS JSON producer runs under your full control

The first important point in the OSIRIS JSON producer documentation is the execution model.

An OSIRIS JSON producer generates a private snapshot of infrastructure at a point in time, it runs from your own environment, uses your credentials or a service accounts to connect to the selected source system or platform and does not require any kind of AI platform, AI agent, MCP server, SaaS provider, intermediary third-party API or expensive consultancies to work.

That boundary matters for technical users: API keys, secrets and raw infrastructure data remain under your full control while the OSIRIS JSON producer discovers inventory and topology data from hyperscalers, cloud and hosting providers, on-premise IT and OT systems.

The OSIRIS JSON producer workflow is deliberate

The OSIRIS JSON Specification describe the OSIRIS JSON producer lifecycle as four steps:

flowchart TB
    n1["Connect to the platform/system<br>and start the discovery"] --> n2["Normalize data into<br>OSIRIS JSON<br>without altering the origin"]
    n2 --> n3["Redaction and removal of<br>secrets and sensitive informations"]
    n3 --> n4["OSIRIS JSON<br>document emission"]

    n1@{ shape: rect}

That sequence matters because each stage has a different responsibility:

• discovery gathers vendor or platform data which output is in the propietary language and format
• normalization maps vendor-specific models into OSIRIS JSON resource types, connections and groups
• redaction strips secrets, credentials and other sensitive configuration values
• emission saves a valid OSIRIS JSON document file

So you get an idea that an OSIRIS JSON document is not as we said before a simple raw export.

The sequence is important, think of finally achieving a result of collecting normalized documents that can be exactly in the same format no matter of the platform your are using.

Security is an OSIRIS JSON producer responsibility

The OSIRIS JSON producer documentation is firm about safety.

OSIRIS JSON producers include built-in guardrails for secret redaction, the --safe-failure-mode flag defaults to fail-closed, preventing credentials and sensitive values from leaking into OSIRIS JSON output. OSIRIS JSON producers also use read-only operations and do not modify device configuration.

There is also no persistent credential cache or producer-side storage beyond the emitted OSIRIS JSON file and the source code is continuously analyzed with SonarCloud and pull requests must pass a quality gate with an A security rating before merge.

That is the right posture. OSIRIS JSON producers sit closest to raw source systems or platforms, which makes them the component that must be most careful with credentials, secrets and configuration data.

OSIRIS JSON producer coverage is incremental

OSIRIS JSON producer coverage is released incrementally.

The current OSIRIS JSON producer documentation lists Cisco and Microsoft Azure as available, it also lists planned work for Amazon Web Services, Arista, Nokia, UniFi, Juniper, Proxmox VE, VMware, Digital Ocean, Leaseweb, and Google Cloud Platform.

Read more on the roadmap page.

Read the OSIRIS JSON producer docs

The producer documentation is available in the producer section of the documentation:

• Read the producer getting-started page
• Read how to install producers
• Read the validation article