Skip to content

VCQA Independent Assessment: Zensical KB Site Stack Page

Date: 2026-07-25 Assessor: Independent AI Agent D Target scope: Stack page assessment only Source target: docs/docs/standards/stacks/zensical-kb-site.md Live target: https://vibecodeqa.online/docs/standards/stacks/zensical-kb-site/ Published assessment URL: https://vibecodeqa.online/docs/standards/assessments/stacks/zensical-kb-site/ Commit: e47a3d44ed97f2b52d7a12a97387796564503ad6

Verdict

partial - This is the strongest of Agent D's three charters. It has a narrow scope, relevant exclusions, credible detection signals, and VCQA-specific concerns around Markdown source, generated output, stable URLs, registry references, and docs drift. It still needs a real rubric before it can judge repos consistently.

Findings

  • medium: docs/docs/standards/stacks/zensical-kb-site.md - No full rubric exists, so scoring and enforcement remain undefined.
  • medium: docs/docs/standards/stacks/zensical-kb-site.md - Rule surface is useful but lacks concrete pass/fail checks for ignored generated output, URL stability, reference provenance, and drift.
  • medium: docs/docs/standards/stacks/zensical-kb-site.md - Detection signals are plausible but should include nav config, broken-link checks, generated site/ policy, and deploy workflow artifacts.
  • low: docs/docs/standards/stacks/zensical-kb-site.md - Benefits list three VCQA surfaces, but this review did not verify their CI/deploy behavior.

Scorecard

Dimension Score Evidence
Scope and exclusions 14/15 Scope is narrow and exclusions are useful.
Composition grounding 13/15 Docs KB and GitHub Actions are the right core items.
VCQA-owned rule surface 15/20 Strong VCQA-specific surface: source of truth, generated output, stable URLs, drift.
Detection and checkability 12/15 zensical.toml, Markdown tree, and deploy workflow are scanner-visible.
Evidence and claim backing 7/10 Docs KB item cites Zensical, Diataxis, ADR, and C4; stack page itself is sparse.
CI/evidence expectations 5/10 Build/deploy workflow is named, but exact gates/artifacts are missing.
Exceptions and anti-gaming 3/10 Generated-site risk is named, but no anti-gaming policy is written.
Maintenance readiness ⅖ Planned status is clear; no review lifecycle.

Total: 71/100

Evidence Reviewed

  • docs/docs/standards/assessment.md
  • docs/docs/standards/stacks/zensical-kb-site.md
  • docs/docs/standards/items/docs-kb.md
  • docs/docs/standards/items/github-actions.md

Claims Not Proven

  • vibecodeqa/docs, vibecodeqa/standards, and the Cloudflare SaaS example docs all satisfy the stated source/build separation.
  • Published URLs are stable in current deployments.
  • Docs drift checks exist where mirrors exist.

Required Fixes

  • Define required CI gates: Zensical build, link check, nav check, generated-output cleanliness, and deploy target evidence.
  • Define exception policy for generated docs, vendored references, archived docs, and deliberately unpublished internal notes.
  • Add concrete examples of failing patterns: committed site/, guessed URLs, stale registry links, and unreferenced standards pages.

Useful Follow-ups

  • Create ref-zensical-kb-site.
  • Promote this stack ahead of VS Code/Tauri if VCQA wants a near-term rubric with strong self-assessment value.