draft: The Policy Set Is the Blueprint — Your Code Was Only Ever Its Derivation

[git-repo-auth]

Draft for the author's review — do not merge. This is the operator's authorial voice; nothing commits or publishes without his pen.

What this is

A public-essay draft for writings/, sequel to the canon principle Bulldoze the App, Keep the Blueprint. Thesis: the durable artifact is the policy set, not the code. Code is a derivation of policy — self-building (regenerate code from the policies) and self-documenting (code cites its policy). The loop: reality rubs the build → the DOLCHEO ledger catches observations / tensions / learnings / constraints / decisions → handoff to policy (new / amended / retired) → builds audited against the whole set → a PRD forms → then the build. The bet: infinite variations of one app at one quality, sharing DNA, because variation lives in the blueprint, not in bespoke code.

• Slug: the-policy-set-is-the-blueprint
• Exposure: draft / public: false (kept off the homepage on purpose)
• ~2,280 words, first person, Socratic register.

Voice/tone derived from

writings/artifacts-are-projections, writings/crew-not-clone, the canon principle bulldoze-but-keep-the-blueprint, plus the corpus-fingerprint rule in odd/ledger/2026-06-13-essay-voice-fingerprint-learning (second-person stakes; question density 7–19 / kwords; ~2,300-word median; reader-as-hero).

Writing-canon gauntlet evidence

• Governing constraints fetched live: writing-canon 6bhvt7, publish-gauntlet wnugeq, triangle-pass pgsg3z, guide-posture pz3eja, dual-context 7riyya, ai-voice-cliches s60mg5.
• Progressive disclosure: stance title; complete-compressed-argument blockquote; ## Summary — extraction key; headers tell the story in sequence; reader as hero.
• Mechanical ghost-writer audit: em dashes 31 → 14 (the one remaining 2-dash line is the See Also footer); negation-parallelism 2 deliberate and far apart; questions 3 → 12; zero formulaic transitions; zero summary clichés; ~2,280 words.
• Not yet run: step 5 challenge (published-essay mode) and step 6 disclosure-by-extraction against the live read model — both belong to your pass, which is why this is filed as a draft.

Honesty flags (please check me)

• "Bulldoze your app" is a canon principle, not a published essay — canon/principles/bulldoze-but-keep-the-blueprint. I treated it as the thematic root and linked it by title. If you meant a different original telling, point me to it.
• The ARS worked example is compressed, and the essay says so in-text. The coordination log has no 2026-07-17 entry (latest seq 1862 @ 2026-07-16T17:23Z); the beats are real but span 07-10 → 07-16, not one afternoon. Anchors: the 07-11 board_import full-replace wipe (the data-loss); the 07-10 provenance-pinned data-model ruling; the "no PRD, no execution" mode-output contract (07-14) + backup/export sweep; and the 07-12 fresh-context validator that restored three dropped pieces pre-merge and confirmed a concurrent write un-clobbered (the gate catch). If a single build-time "data-loss blocked at the gate" episode exists that I missed, I'll re-anchor.
• Read hardest here: whether the ARS section sounds like you telling it vs. me reporting it.

[github-actions]

Canon Quality — Homepage Surfacing ℹ️

50 essay(s) scanned. Soft report — never blocks; the hard field gate is the Frontmatter Schema job.

1 essay(s) are NOT on the homepage feed (confirm this is intentional):

Essay	Surface
writings/the-policy-set-is-the-blueprint.md	hidden — public=False — NOT a published essay; will not surface publicly

To promote to the homepage: set public: true and exposure: public.

_{Report: scripts/surfacing-report.py · Canon: klappy://canon/constraints/frontmatter-validation-before-merge}

[github-actions]

Canon Quality — Frontmatter Schema ✅

All 50 file(s) in writings/ conform to klappy://canon/meta/frontmatter-schema.

_{Validator: scripts/validate-frontmatter.py · Canon: klappy://canon/constraints/frontmatter-validation-before-merge · Run: #375}

[github-actions]

Canon Quality — oddkit_audit ✅

No dead klappy:// references or legacy link patterns found in writings/. 51 files scanned.

_{Spec: klappy://docs/oddkit/specs/oddkit-audit · Workflow: .github/workflows/canon-quality.yml · Run: #375}

[github-actions]

Canon Quality — P0010 Retrieval-Readiness ⚠️

Soft report for klappy://canon/constraints/retrieval-disclosure-contract. 700 files scanned. Never blocks — informational until the corpus is ready to enforce.

• Blocking-class findings: 15 (structural fields the contract would filter on)
• Warnings: 0 (kind resolves to unknown)
• Informational: 13 (exempt templates/archive/drafts)

Kind distribution: {'essays': 52, 'canon': 240, 'apocrypha': 38, 'docs': 304, 'journals': 60, 'unknown': 6}
Kind source: {'path': 565, 'frontmatter': 129, 'none': 6} (frontmatter-primary, path-secondary)
Default-include visibility: 596 visible, 104 hidden (journals/apocrypha/unknown)

By rule: {'audience-invalid': 2, 'exposure-missing': 5, 'tier-missing': 5, 'tier-invalid': 7, 'fm-missing': 3, 'kind-unresolvable': 6}

These are not schema violations (see the Frontmatter Schema job for those on writings/). They are corpus-readiness signals for the retrieval contract: invalid/missing audience, exposure, tier, and docs whose kind cannot be resolved. Fix in a corpus-cleanup PR before the contract flips to enforcing. See the retrieval-readiness-findings artifact for the full list.

_{Validator: scripts/audit-retrieval-readiness.py · Constraint: klappy://canon/constraints/retrieval-disclosure-contract · Run: #375}