Skip to content

A Day in the Life: Docs-as-Code Personas

Personas: IA (Information Architect), CS (Content Strategist), DQR (Documentation Quality Reviewer), ADS (Automated Doc System Engineer)

Morning: Architecture and Strategy

IA reviews the documentation structure for the latest release. New features need new pages, existing pages need updates, and deprecated features need archival. IA designs the information architecture: navigation hierarchies, cross-reference links, and content groupings that help users find what they need in three clicks or fewer.

IA produces a sitemap update showing new pages, modified pages, and deprecated pages. Each page has a defined audience, purpose, and relationship to other pages in the documentation tree.

CS defines the content strategy for the release. CS determines the narrative arc: what story does this release tell? What are the key messages for each audience (developers, operators, executives)? CS creates content briefs for each new or updated page, specifying tone, length, key points, and required code examples.

Midday: Automation

ADS maintains the automated documentation pipeline. Every commit triggers a documentation build: Jinja2 templates render Markdown pages, code examples are extracted and validated, cross-references are checked, and broken links are flagged. ADS configures the CI/CD pipeline to fail the build on documentation errors, treating docs with the same rigor as code.

ADS also manages the template system -- 21 Jinja2 templates that generate persona pages, action guides, cross-reference matrices, and ecosystem prompt pages. When a new persona is added to the YAML registry, ADS's pipeline automatically generates all 56 documentation files.

Afternoon: Quality Review

DQR reviews the generated documentation for quality. DQR checks for completeness (all R.I.S.C.E.A.R. sections present), accuracy (code examples actually run), consistency (terminology matches the glossary), and readability (sentence length, passive voice usage, jargon density). DQR uses a structured rubric and scores each page on a 5-point scale.

DQR provides feedback to CS and IA, completing the documentation quality loop. Pages that score below the threshold are returned for revision with specific improvement recommendations.

Tools Used

  • DocGenerator for automated page generation
  • Jinja2 templates from src/fcc/templates/docs/
  • fcc validate-docs for automated quality checks
  • fcc sitemap for navigation structure generation

Key Outputs

  • Updated sitemaps and navigation hierarchies (IA)
  • Content briefs and narrative strategies (CS)
  • Automated documentation pipelines (ADS)
  • Quality review reports with scores and recommendations (DQR)