reversa-docs

To generate an offline HTML documentation mini-site, this Skill orchestrates mapping, analysis, storytelling, and publishing into a self-contained package within `.reversa/documentation/`. It bundles vendor assets for `file://` compatibility and omits pages whose required sources are absent.

Reproducible visuals for architectural documentation are built using a deterministic seed derived from `.reversa/soul.md` or the project name. This approach ensures visual rendering remains stable across multiple runs, and you can override the default seed using the `--seed` parameter.

You need SDD specs to generate specific project documentation pages such as architectural visualizations, metric dashboards, and per-feature pages. If core sources are missing, the pipeline applies a graceful greenfield fallback to produce a minimal site instead of failing.

You run isolated regeneration for focused updates by triggering the sequential specialist pipeline with targeted parameters. The pipeline processes Mapper, Analyst, Storyteller, and Publisher in a fixed order, persisting state in `.reversa/documentation/.state.json` to support selective rebuilds.

Your documentation mini-site omits specific pages like metrics or glossaries when their required sources are absent from `.reversa/state.json`. The orchestration pipeline automatically detects missing core knowledge and skips the generation of those corresponding pages.

You cannot use this documentation orchestration pipeline without reading `.reversa/state.json`, which is a strict requirement for extracting core knowledge. An optional `.reversa/documentation/.config.json` file can also be read to override default generation settings.