You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
As a team adopting pair (docs reader / tech lead evaluating what to turn on) I want one docs-site reference page that catalogs every opt-in quality configuration with a uniform schema — and explicitly flags what is instead mandatory/default-on So that I discover, in one place, which quality mechanisms exist, their default state, and how to enable each — instead of hunting across concepts/, customization/, and reference/.
Where: docs-site (apps/website), a new page under reference/ — content only, no framework/skill logic.
Refined: Story is detailed, estimated, and ready for development
In Progress: Story is actively being developed
Done: Story delivered and accepted
Classification
risk:green · cost:green · coupling: not assessed — no domain artifacts
Matrix — per dimension
Dimension
Tier
Source
Note
Service/domain criticality
green
KB default
docs-site content; no runtime/service surface
Change/diff risk
green
story scope
one new MDX page + nav entry + a few cross-link edits, content-only
Business impact
green
subdomain class (supporting)
documentation; no user-facing product behavior
Security relevance
green
path heuristic
pure MDX under apps/website/content/docs, no secrets/auth/infra paths
Coupling balance
not assessed
absent
no domain artifacts for this scope; excluded from tier max (D21/D38)
Risk tier = max across assessed dimensions = 🟢 green. Cost class = green (no build/runtime/infra cost — a docs page). risk:* projection is active in this repo, so risk:green is emitted; cost class shown for completeness.
Acceptance Criteria
Functional Requirements
Given-When-Then Format:
Given a reader on the docs-site who wants to know which quality mechanisms are configurable When they open the new "Quality gates & configuration" reference page Then they see a single catalog listing every opt-in quality knob, each row carrying the uniform schema: what it is · default state · adoption flag / mechanism · how to enable · link to owning guideline/concept page.
Given the catalog page When the reader scans it Then it covers at least: Pre-merge tiering (Pre-merge tiering: disabled default, way-of-working.md), Coverage guardrail (Coverage guardrail: disabled default, way-of-working.md + tech/coverage-baseline.md), Template override (file-existence mechanism adoption/tech/templates/<name>-template.md), Tag Projection (## Tag Projection in tech/risk-matrix.md; risk proposed, cost/others opt-in), and Criticality Table / Overrides (tech/risk-matrix.md adoption delta).
Given the same page When the reader looks for what is NOT optional Then a clearly separated section lists the mandatory / default-on gates — with deterministic secret-scanning (gitleaks, R6.5/D24) called out as always-active and never tier-scoped — so the opt-in vs always-on distinction is unambiguous.
Given the docs navigation and the existing feature pages When the reader browses the reference/ section or lands on a related page (concepts/tag-driven-gates, reference/quality-model, customization/templates) Then the new page is listed in the reference/ nav (meta.json) and linked from reference/index.mdx, and each related feature page cross-links to it as the unifying index.
Given the page content When each opt-in knob's default state and enable-flag is stated Then it matches the current source of truth (e.g. framework default disabled for tiering and coverage guardrail, per the KB/dataset template — not pair's own project-level opt-in), and each row deep-links to its owning concept/reference/guideline page rather than duplicating that page's prose.
Business Rules
Uniform schema is mandatory for every catalog row: what · default · flag/mechanism · how to enable · guideline link. No row omits a column.
The page is an index/catalog, not a re-explanation: it links to the owning page (tag-driven-gates, quality-model, templates) for depth; it does not duplicate their content.
Defaults quoted are the framework/dataset defaults (default-off for tiering & coverage guardrail), explicitly distinguished from pair's own dogfooding opt-ins where relevant.
Content only — the story introduces no skill, CLI, or pipeline logic; it documents existing mechanisms.
Edge Cases and Error Handling
Invalid Input: N/A — static docs page; no user input.
System Errors: build-time — page must pass the docs-site build (MDX/lint, internal-link check); broken internal links fail the build.
Boundary Conditions: a knob whose owning page does not yet exist → link to the closest existing owner (concept or quality-model) and note it, rather than a dead link.
Exceptional Scenarios: a new opt-in knob ships later → page carries an explicit maintenance note (this catalog is the unifying index; each feature PR adds its row here as part of its DoD), so staleness is visible.
Internal links resolve (docs-site link check passes)
Nav updated: page added to reference/meta.json and linked from reference/index.mdx
Cross-links added from related feature pages to the catalog
Docs-site build + lint pass
Quality Assurance
Every catalog row carries the full uniform schema (no missing column)
Opt-in vs mandatory/default-on distinction present and correct (secret-scan called out)
Defaults/flags match current source of truth (way-of-working flags, risk-matrix sections, template resolution)
No content duplication — rows link to owning pages
Renders correctly (local docs-site preview)
Deployment and Release
Merged to main; docs-site deploy (Vercel) succeeds
Page reachable in production nav under Reference
Story Sizing and Sprint Readiness
Refined Story Points
Final Story Points: S (2) Confidence Level: High Sizing Justification: One new MDX page with a well-understood schema (mechanisms already documented on their own pages), plus a nav entry, an index link, and a handful of one-line cross-link edits. No logic, no tests beyond the docs build/link check. Breadth (5 knobs + mandatory-gates contrast) is content assembly, not complexity.
Sprint Capacity Validation
Sprint Fit Assessment: Fits comfortably in a single sprint. Development Time Estimate: ~0.5 day Testing Time Estimate: ~0.25 day (build + link check + preview review) Total Effort Assessment: Fits within sprint capacity: Yes
Story Splitting Recommendations
Not required — story is already small and delivers end-to-end value (a reachable, linked catalog page) in one slice.
Dependencies and Coordination
Story Dependencies
Prerequisite Stories: None hard-blocking. Content is most accurate once the opt-in reframes it documents are settled — coverage guardrail opt-in (#282) and pre-merge tiering (#258, already opt-in). If any default is still in flux, quote the current KB/dataset default and update on change (maintenance note covers this). Dependent Stories: None. Shared Components: Sibling of #353 (operational narrative "how the delivery process changes") — complementary, not blocking. This story is the configuration reference (knob catalog); #353 is the process narrative. Distinct from reference/quality-model (conceptual model) and reference/configuration (registry config.json).
Testing Methods: Docs-site build + MDX lint + internal-link check; manual preview review against the AC checklist (schema completeness per row, opt-in vs mandatory split, cross-links present). Test Data Requirements: None. Environment Requirements: Local docs-site dev/build; production deploy verification.
User Validation
User Feedback Collection: Review by a maintainer familiar with the quality model; optional adopter read-through. Success Metrics: A reader can, from this one page, name every opt-in knob, its default, and how to enable it — without opening the source repo. Rollback Plan: Revert the docs PR (content-only; no runtime impact).
Notes and Additional Context
Refinement Session Insights: Story reframed from the Draft "knobs" brief into a refined DoR-complete spec. Scope confirmed content-only on apps/website. Parent epic set to #93. Team Concerns: Keep the catalog an index (link, don't duplicate) so it does not drift from the owning pages; the maintenance note makes staleness visible and ties row-upkeep to each feature PR's DoD. Future Considerations: A staleness/link-check gate could later assert every opt-in flag documented in way-of-working.md/risk-matrix.md has a catalog row (out of scope here). Documentation Links: concepts/tag-driven-gates.mdx, reference/quality-model.mdx, customization/templates.mdx, reference/configuration.mdx (registry config — explicitly the other configuration page).
Technical Analysis
Implementation Approach
Technical Strategy: Add one new MDX reference page and wire it into the docs-site nav; add cross-links from the related feature pages. No framework, skill, CLI, or pipeline changes. Purely apps/website docs content. Key Components:
Nav: apps/website/content/docs/reference/meta.json (add the page) and apps/website/content/docs/reference/index.mdx (link under "Structure & configuration").
Cross-links: one line each in concepts/tag-driven-gates.mdx, reference/quality-model.mdx, customization/templates.mdx. Data Flow: N/A — static content rendered by the existing Fumadocs/Next.js docs pipeline. Integration Points: Docs-site build + internal-link validation; Vercel deploy. Design: not required — approach is understood; standard docs page following existing reference/ page conventions.
All internal links must resolve (build fails otherwise).
Content matches current source of truth for defaults/flags.
Accessibility/readability per existing docs-site standards (no new components required).
Technical Risks and Mitigation
Risk
Impact
Probability
Mitigation Strategy
Documented defaults drift from KB/dataset source of truth
Medium
Medium
Quote current defaults, deep-link owning pages, add explicit maintenance note tying row-upkeep to each feature PR's DoD
Content duplicates owning pages and diverges
Low
Medium
Catalog is an index — rows link out, do not restate depth
Broken internal links
Low
Low
Docs-site build/link check in CI
Spike Requirements
Required Spikes: None. Estimated Spike Effort: 0.
Task Breakdown
T1 — Draft catalog page: create apps/website/content/docs/reference/quality-gates-configuration.mdx with frontmatter and the uniform-schema table(s) covering pre-merge tiering, coverage guardrail, template override, tag projection, criticality table/overrides (AC1, AC2, AC5).
T2 — Mandatory/default-on contrast section: add the "what is NOT optional" section, secret-scanning (gitleaks, R6.5/D24) called out as always-active (AC3).
T3 — Maintenance note: add the staleness/maintenance note tying each future knob's row to its feature PR DoD (edge case, business rule).
T4 — Nav wiring: add the page to apps/website/content/docs/reference/meta.json and link it under "Structure & configuration" in apps/website/content/docs/reference/index.mdx (AC4).
T5 — Cross-links: add one cross-link to the catalog from each of concepts/tag-driven-gates.mdx, reference/quality-model.mdx, customization/templates.mdx (AC4).
T6 — Verify: run docs-site build + internal-link check, preview-review against the AC checklist (schema completeness, opt-in vs mandatory split, links resolve).
Refinement Completed By: Product Engineer (AI-assisted refinement) Refinement Date: 2026-07-23 Review and Approval: Pending product-owner confirmation on board move to Refined
Story Statement
As a team adopting pair (docs reader / tech lead evaluating what to turn on)
I want one docs-site reference page that catalogs every opt-in quality configuration with a uniform schema — and explicitly flags what is instead mandatory/default-on
So that I discover, in one place, which quality mechanisms exist, their default state, and how to enable each — instead of hunting across
concepts/,customization/, andreference/.Where: docs-site (
apps/website), a new page underreference/— content only, no framework/skill logic.Epic Context
Parent Epic: #93 Website & Documentation — Landing + Docs + Brand Identity
Status: Refined
Priority: P1 (Should-Have)
Status Workflow
Classification
risk:green·cost:green· coupling: not assessed — no domain artifactsMatrix — per dimension
apps/website/content/docs, no secrets/auth/infra pathsRisk tier = max across assessed dimensions = 🟢 green. Cost class = green (no build/runtime/infra cost — a docs page).
risk:*projection is active in this repo, sorisk:greenis emitted; cost class shown for completeness.Acceptance Criteria
Functional Requirements
Given-When-Then Format:
Given a reader on the docs-site who wants to know which quality mechanisms are configurable
When they open the new "Quality gates & configuration" reference page
Then they see a single catalog listing every opt-in quality knob, each row carrying the uniform schema: what it is · default state · adoption flag / mechanism · how to enable · link to owning guideline/concept page.
Given the catalog page
When the reader scans it
Then it covers at least: Pre-merge tiering (
Pre-merge tiering: disableddefault,way-of-working.md), Coverage guardrail (Coverage guardrail: disableddefault,way-of-working.md+tech/coverage-baseline.md), Template override (file-existence mechanismadoption/tech/templates/<name>-template.md), Tag Projection (## Tag Projectionintech/risk-matrix.md;riskproposed,cost/others opt-in), and Criticality Table / Overrides (tech/risk-matrix.mdadoption delta).Given the same page
When the reader looks for what is NOT optional
Then a clearly separated section lists the mandatory / default-on gates — with deterministic secret-scanning (gitleaks, R6.5/D24) called out as always-active and never tier-scoped — so the opt-in vs always-on distinction is unambiguous.
Given the docs navigation and the existing feature pages
When the reader browses the
reference/section or lands on a related page (concepts/tag-driven-gates,reference/quality-model,customization/templates)Then the new page is listed in the
reference/nav (meta.json) and linked fromreference/index.mdx, and each related feature page cross-links to it as the unifying index.Given the page content
When each opt-in knob's default state and enable-flag is stated
Then it matches the current source of truth (e.g. framework default
disabledfor tiering and coverage guardrail, per the KB/dataset template — not pair's own project-level opt-in), and each row deep-links to its owning concept/reference/guideline page rather than duplicating that page's prose.Business Rules
Edge Cases and Error Handling
Definition of Done Checklist
Development Completion
reference/meta.jsonand linked fromreference/index.mdxQuality Assurance
Deployment and Release
Story Sizing and Sprint Readiness
Refined Story Points
Final Story Points: S (2)
Confidence Level: High
Sizing Justification: One new MDX page with a well-understood schema (mechanisms already documented on their own pages), plus a nav entry, an index link, and a handful of one-line cross-link edits. No logic, no tests beyond the docs build/link check. Breadth (5 knobs + mandatory-gates contrast) is content assembly, not complexity.
Sprint Capacity Validation
Sprint Fit Assessment: Fits comfortably in a single sprint.
Development Time Estimate: ~0.5 day
Testing Time Estimate: ~0.25 day (build + link check + preview review)
Total Effort Assessment: Fits within sprint capacity: Yes
Story Splitting Recommendations
Not required — story is already small and delivers end-to-end value (a reachable, linked catalog page) in one slice.
Dependencies and Coordination
Story Dependencies
Prerequisite Stories: None hard-blocking. Content is most accurate once the opt-in reframes it documents are settled — coverage guardrail opt-in (#282) and pre-merge tiering (#258, already opt-in). If any default is still in flux, quote the current KB/dataset default and update on change (maintenance note covers this).
Dependent Stories: None.
Shared Components: Sibling of #353 (operational narrative "how the delivery process changes") — complementary, not blocking. This story is the configuration reference (knob catalog); #353 is the process narrative. Distinct from
reference/quality-model(conceptual model) andreference/configuration(registryconfig.json).Team Coordination
External Dependencies
Third-party Integrations: None.
Infrastructure Requirements: Existing docs-site build + Vercel deploy.
Compliance Requirements: None.
Validation and Testing Strategy
Acceptance Testing Approach
Testing Methods: Docs-site build + MDX lint + internal-link check; manual preview review against the AC checklist (schema completeness per row, opt-in vs mandatory split, cross-links present).
Test Data Requirements: None.
Environment Requirements: Local docs-site dev/build; production deploy verification.
User Validation
User Feedback Collection: Review by a maintainer familiar with the quality model; optional adopter read-through.
Success Metrics: A reader can, from this one page, name every opt-in knob, its default, and how to enable it — without opening the source repo.
Rollback Plan: Revert the docs PR (content-only; no runtime impact).
Notes and Additional Context
Refinement Session Insights: Story reframed from the Draft "knobs" brief into a refined DoR-complete spec. Scope confirmed content-only on
apps/website. Parent epic set to #93.Team Concerns: Keep the catalog an index (link, don't duplicate) so it does not drift from the owning pages; the maintenance note makes staleness visible and ties row-upkeep to each feature PR's DoD.
Future Considerations: A staleness/link-check gate could later assert every opt-in flag documented in
way-of-working.md/risk-matrix.mdhas a catalog row (out of scope here).Documentation Links:
concepts/tag-driven-gates.mdx,reference/quality-model.mdx,customization/templates.mdx,reference/configuration.mdx(registry config — explicitly the other configuration page).Technical Analysis
Implementation Approach
Technical Strategy: Add one new MDX reference page and wire it into the docs-site nav; add cross-links from the related feature pages. No framework, skill, CLI, or pipeline changes. Purely
apps/websitedocs content.Key Components:
apps/website/content/docs/reference/quality-gates-configuration.mdx(frontmatter + uniform catalog table(s) + mandatory/default-on contrast section + maintenance note).apps/website/content/docs/reference/meta.json(add the page) andapps/website/content/docs/reference/index.mdx(link under "Structure & configuration").concepts/tag-driven-gates.mdx,reference/quality-model.mdx,customization/templates.mdx.Data Flow: N/A — static content rendered by the existing Fumadocs/Next.js docs pipeline.
Integration Points: Docs-site build + internal-link validation; Vercel deploy.
Design: not required — approach is understood; standard docs page following existing
reference/page conventions.Technical Requirements
title/description, heading hierarchy, tables).Technical Risks and Mitigation
Spike Requirements
Required Spikes: None.
Estimated Spike Effort: 0.
Task Breakdown
apps/website/content/docs/reference/quality-gates-configuration.mdxwith frontmatter and the uniform-schema table(s) covering pre-merge tiering, coverage guardrail, template override, tag projection, criticality table/overrides (AC1, AC2, AC5).apps/website/content/docs/reference/meta.jsonand link it under "Structure & configuration" inapps/website/content/docs/reference/index.mdx(AC4).concepts/tag-driven-gates.mdx,reference/quality-model.mdx,customization/templates.mdx(AC4).Refinement Completed By: Product Engineer (AI-assisted refinement)
Refinement Date: 2026-07-23
Review and Approval: Pending product-owner confirmation on board move to Refined