3. Context and Scope¶
HermesBaby is the system-of-interest. It operates inside an engineering organisation's development environment and interacts with engineers, version-control infrastructure, CI pipelines, rendering tools, and publishing targets.
Fig. 3.1 HermesBaby in its operational context: engineers, Git, CI, rendering toolchain, documentation portal, and access-control mechanism.¶
3.1. External Actors and Systems¶
Actor / System |
Interaction with HermesBaby |
|---|---|
Software / Systems Engineer |
Invokes |
Git Repository |
HermesBaby reads documentation source ( |
CI Pipeline |
An automated pipeline (e.g. GitHub Actions) triggers |
Sphinx / MyST-Parser Toolchain |
HermesBaby invokes Sphinx with MyST-Parser and a project-specific set of Sphinx extensions. Sphinx reads the source tree, resolves cross-references and citations, renders diagrams, and writes the output (HTML, optionally PDF) to the configured output directory. |
PlantUML |
The |
draw.io CLI |
The |
Documentation Portal / Web Server |
Receives the static HTML output published by |
Access-Control Mechanism |
HermesBaby generates access-control artefacts (e.g. |
AI Assistant / LLM |
Engineers use large language model assistants to synthesise information from standards, specifications, and codebases, and to generate draft documentation content. Because HermesBaby source files are plain text stored in Git, they are directly indexable and referenceable by AI tools — stable headings and cross-references act as anchors for precise, scoped answers [MW25]. AI-assisted authoring is a natural extension of the Docs-as-Code workflow. |