Vedic remaining work register

Status: Active documentation register; implementation requires a separately authorized work package. Last reconciled: 12 October 2026. Baseline: engine main, version 6.9.9. Sections 36–38 retain the source-published adversarial repair checkpoint; sections 39–44 record the selected MC marriage delivery and both published performance passes. Section 45 preserves the four-area research checkpoint. Sections 46–47 record the selected housing/Vastu, commerce, travel and VV/KP marriage implementation and source publication. Section 48 records VED-017's bounded Gochar windows and final local engine/facade/REST acceptance. Section 51 closes the selected 21-entry VED-012 yoga package across engine/facade/REST. Section 54 closes VED-013's selected three Chara recipes and three Argala assessments locally. Section 59 closes VED-014's named Santhanam/BPHS Kalachakra profile locally. Sixteen of 24 stable feature packages have reached their selected delivery checkpoints; eight remain open/bounded. Release and deployment remain separate. Scope: Moira engine, curated Python surfaces, REST contracts, source evidence and validation. Website and Urania adoption are separate product work.

This is the current index of remaining Vedic work. It supersedes the Vedic backlog claims in the original completion roadmap, Phase 2 gap register, paused Tier 2 tracker, the Vedic entries in the frontiers register, and section 6.2 of the August coverage audit. Those records remain historical evidence, including their source leads and earlier pauses.

The 6 October surface receipt records the completed export/Gochar REST package. It is a completion receipt, not the remaining-work list. Source presence, public export, REST exposure, source validation, release and deployment are distinct states.

Current extensions: VED-015 is committed and pushed in engine 0b34f22841eb6b5f7dace9128b4d01cc11c480d3 and wiki 4c12d88b9825d562be38f7d3b67c37a703449735. Its source, policy, engine/REST and verification receipt records the admitted daily contract. Source publication is not deployment. The user subsequently selected lunar-month and festival rules; VED-023/024 below preserve their distinct admission and remaining-work boundaries.

1. Status and priority

Status	Meaning
OPEN_INTEGRATION	A calculation or policy already exists, but a public, chart-composition or REST surface omits it. Admission still requires appropriate bounds and evidence.
OPEN_ENGINE	A concrete behavior or helper contract remains incomplete in current code.
SOURCE_RESEARCH	A recorded extension needs edition collation, named lineage, typed results and fixtures before implementation. An old source citation does not admit a rule.
EVIDENCE_REVIEW	Existing code/evidence must be reconciled before a broader completion claim is justified.
CANDIDATE	Optional expansion recorded for a scope decision; not a committed delivery obligation.
BOUNDED_ADMISSION	A finite contract is implemented, while named exceptional rules or authority reconciliation remain open.
LOCAL_COMPLETE	Authorized engine/public/REST package implemented and validated locally; publication, release and adoption remain distinct.

P1 closes gaps in existing products; P2 deepens existing families or provides the next foundational composition; P3 is a larger separately scoped programme. These are sequencing suggestions, not effort estimates, dates or an instruction to resume paused work. No item must mechanically pass twelve identical phases: use the constitutional gates that apply to its actual object.

There are 24 stable numbered work packages below: 8 remain open/bounded, including bounded VED-023. VED-001/002/003/004/005/006/007/008/009/010/011/012/013/014/015/017 reached their selected feature-delivery checkpoints; VED-001/002/003/004/005/006/007/008/009/010/011/015 are in the earlier engine/wiki source-publication history; VED-017's authorized publication package is recorded in section 49. The separate repair acceptance in sections 36–37 closes all sixteen adversarial findings, including VED-010's optional Shadbala composition failure; it does not change these feature-delivery counts. VED-017's bounded modern window closure adds one locally complete package in section 48. Several contain source-separated subquestions rather than a promise to implement every tradition. Five optional candidates follow in a separate table. Shared source/REST/validation requirements are not counted as additional techniques.

2. Existing engine and REST closure

ID	Work still required	Current evidence	Status / priority
VED-001	Locally complete for strict one/two-cycle selection, exact seven-body/optional node-pair inputs and canonical execution receipts. Complete-cycle/co-lord source collation, alternative formulations and later cycles remain separate VED-013/021 research.	Chara cycle standard, approved plan and validation receipt. Default period arithmetic retained; published second-cycle evidence is explicitly partial.	LOCAL_COMPLETE; engine/wiki source-published; release/deployment separate
VED-002	Locally complete for the collated BPHS within-sign Navamsa-ordinal profile: exact partition ownership, strict body/Moon/name/ghati inputs, canonical trace, optional chart/node and sunrise-owned birth composition, curated Python/facade and typed REST parity. Degree-based traditions and full legacy-effect prose certification remain separately sourced scopes.	Source audit, executed plan, admission standard and validation receipt: 278 distinct latest passes, zero disagreements in 14,580 independent rational cases, corrected Sa=4 source fixture, real reader/startup and direct/HTTP parity. Legacy effects explicitly retain uncertified, unevaluated provenance.	LOCAL_COMPLETE; engine/wiki source-published at the section 25 baseline; release/deployment separate
VED-003	Locally complete for transport, four edition-owned names, Santhanam signs, modern composed full positions and the explicit classical_derived_linear full-degree policy through engine/facade/REST/strength. The wider library establishes classical/commentarial support at the reviewed scopes. Direct classical D60 degree prescription and exact primary-published planetary pairs are optional future research outside this closed implementation scope.	D60 standard, classical-derived admission and execution receipt. Modern and derived-classical provenance stay distinct; frozen witnesses and rational boundary checks validate both selected mappings. Harmonic defaults retained.	LOCAL_COMPLETE; engine/wiki source-published; release/deployment separate
VED-004	Locally complete: seven weights validated/exposed, actual selected/applied/reserved policy receipts, and personal-score frame precedence reconciled.	Personalized Muhurta standard and validation receipt. Reserved classical-rule flag is fixed True; no alternate school is silently selected.	LOCAL_COMPLETE; engine/wiki source published, release/deployment separate
VED-005	Locally complete for the selected 30 Vimshottari, alternate-dasha, Shadbala, dignity, Sade Sati and Vedic-profile routes: strict inputs/identities, admitted period trees, selected-policy preflight, serving-reader repair, consistent alternate chart frames and applied-policy receipts. Additional range/execution/response budgets require separately selected operational policy.	Admission standard and audit/repair validation receipt: 175 distinct latest passing tests, real discovered/configured-kernel lifecycles and explicit compatibility changes. Earlier Panchanga/Muhurta hardening remains its original scope; neither package claims a whole-Vedic numerical certification.	LOCAL_COMPLETE; engine/wiki source published, exact SHAs in section 21; release/deployment separate

VED-001 is locally complete. VED-003 admits transport, selected-edition names, source-owned signs, modern composed full positions and a usable classical-derived full-degree profile. Both full source profiles have completed computational validation, including independent software witnesses and sixteen real-reader date/frame chart batches. Classical support is established at the separately documented scopes; direct classical D60 degree prescription and exact published planetary pairs remain unclaimed source-evidence frontiers, not implementation gates. A single rounded arudha-point witness is conditional. VED-002 now closes the selected BPHS numeric/state, name/clock, birth composition and public/REST contract. Its receipt separates source formula and transport validation from legacy conditional-effect prose certification and predictive claims.

3. Muhurta work recovered from the paused notes

ID	Work still required	Current evidence and boundary	Status / priority
VED-006	Locally complete for natal-aware sampled composition: actual personal evaluator, explicit true ayanamsa and serving-reader clocks, visible failures, threshold-qualified consecutive runs, ranked peak/bracket evidence and operational caps, with bounded REST search. Exact transitions and sunrise-owned search would require separately selected composition.	Search engine, repaired legacy scorer/tuple adapters, eight shared root/facade/Vedic exports, two facade methods and POST /v1/muhurta/search; owning standard and validation receipt. Existing rule profile retained; no Western doctrine or new classical rule admitted.	LOCAL_COMPLETE; bounded sampled/JD-weekday contract, engine/wiki source published, release/deployment separate
VED-007	Complete locally: Abhijit, Brahma, Godhuli, Vijaya, selected Amrita, Ravi Yoga and Sarvarthasiddhi; named source policies, precise solar/phase boundaries, source exceptions, uncertainty and partial states, public/facade and five total REST routes. Amrita Siddhi and Kalaprakasika Amirtha are separate selectable identities.	Five-name source decision, standard, 383-test completion receipt. All five formerly outstanding names implemented and validated, including both startup reader paths and real DE441. Generic scoring/search and activity suitability remain separate products.	LOCAL_COMPLETE; all-seven-family implementation validated and included in this engine/wiki source-publication package; release/deployment separate
VED-008	Locally complete for named Panchaka, alternative MC/PS Tara-cycle, nine-Yoga timing and eleven-Karana/Bhadra assessments, including solved sunrise-day cells. Existing Tara/Chandra, scoring and search remain compatible. Specific excluded readings, polar discontinuous Lagna timing and general score integration are separately bounded.	Source research, standard and 622-test validation receipt. Sixteen curated exports, three facade methods, three typed REST routes; source conflicts and derived clocks remain explicit.	LOCAL_COMPLETE; included in this authorized engine/wiki source-publication package; release/deployment separate
VED-009	Locally complete finite six-detector/three-exception package: source-specific stellar Vishanadi, both Yamaghantas and temporal nakshatra/tithi/Lagna Gandanta; raw evidence retained after opt-in neutralization. Other Tyajya types, angular/Abhukta variants and the complete twenty-one-dosha catalogue remain explicitly excluded extensions.	Source packet, standard and validation receipt. Fifteen owning exports, three facade methods and three strict REST routes; full-parent events, numerical bands, carryover and source-scoped cancellation evidence. Strength-dependent exceptions belong to VED-010.	LOCAL_COMPLETE and published at engine 571788a / wiki ed68937; broader listed extensions remain unadmitted
VED-010	Source-selected general/marriage Lagna components, both Navamsa readings, quarter-aspect support, MC placement score, raw restrictions and optional MC88 exceptions. Canonical Shadbala remains inspectable context; its dated multi-war failure and subsequent context/clock admission defects are repaired.	Source record, standard, feature validation, combined repair validation. Eleven curated exports, three facade methods and three REST routes; source-backed D9 and Venus/Moon table corrections in shared owners.	LOCAL_COMPLETE; feature source-published and all identified repairs included in the section 38 publication package; broader listed remedies remain excluded
VED-011	Selected MC ordinary/personal/Godhuli marriage and both performance passes remain source-published. MC48 residential housing and ordinary land travel, MC04 purchase/sale/first opening, named KP travel components and VV/KP ordinary marriage now complete the selected four-area engine/facade/REST package locally. Further rituals, commercial acts and source traditions retain explicit extension boundaries.	Purpose standard, completed plan and final acceptance. Source manifests, conditional evidence, strict transport, bounded real windows, exact MC regression and operational limits are recorded.	LOCAL_COMPLETE for the selected package; engine/wiki source-publication package recorded in section 47; release/deployment separate

The May tracker is retained as a research trail. Its historical "fully wired" claim did not establish the current contract. VED-009 now has its own validated dosha/Parihara owner; the legacy classification has not acquired that richer contract. VED-010 Lagna-strength composition is locally complete for its named policies and instantaneous contract.

4. Other established-family depth and foundational additions

ID	Work still required	Current evidence and admission gate	Status / priority
VED-012	Locally complete for the selected 21 source entries: separate JP 7.14 and Raman no. 252 Navamsa profiles, fourteen BPHS wealth rules and five selected royal entries. Legacy admission, missing-motion evidence and applied-policy transport repairs are also complete. BPHS 39.35 needs source resolution of aspect targets/type; broader chapter-specific and unidentified Raman source leads remain separate extensions.	Standard, executed plan and validation receipt: 573 passing selected tests, 44 independent source cases across all public/HTTP paths, 144-chart regression and bounded resource-free cost. See sections 51-52 for closure and source publication; the former global docstring findings are resolved.	LOCAL_COMPLETE; authorized engine/wiki source-publication package in section 52; release/deployment separate
VED-013	Locally complete for three Rangacharya Chara recipes, separate optional calendar projection and BPHS/PVR/Rath Argala assessments. Existing Chara and valid legacy Argala arithmetic are preserved; direct-input and lineage/receipt repairs are complete. Full BPHS/PVR Chara, Narayana, Yogada and general Argala strength retain explicit admission gates.	Source packet, standard, executed seven-stage plan and validation receipt: 951 passing selected tests after three adversarial repairs, independent printed tables and boundary arithmetic, all public/HTTP surfaces and bounded cost. Source conflicts remain visible. See sections 54–57.	LOCAL_COMPLETE for the selected six profiles; original engine/wiki publication in section 55; validated repair publication in section 57; release/deployment separate
VED-014	Locally complete for bphs_santhanam_longitude_row_cycle_v1: exact source arithmetic, nine occurrence-aware MDs and optional full-parent ADs, explicit calendar clocks, reader-bound Moon composition, bounded active lookup and equally capable Python/facade/REST. Competing traditions, deeper levels, full gati taxonomy and eligibility evaluation remain separate capabilities.	Source research, executed seven-stage plan, backend standard, API and validation receipt: 249 feature + 545 regression tests passed, including real DE441 and HTTP parity.	LOCAL_COMPLETE; publication/release/deployment separate
VED-015	Completed local daily contract: local date/zone/location, sunrise-owned Vara, retained sunrise/sunset crossings, solved angular limb endings, repeated/skipped sunrise evidence, half-open coverage, and unavailable solar states.	Daily engine, curated root/facade/Vedic exports and POST /v1/panchanga/day; source/policy/validation standard. Three named sunrise conventions and existing ayanamsa choices are explicit. Lunar months/festivals remain separate scope.	LOCAL_COMPLETE; engine/wiki source published, release/deployment separate
VED-016	Source and admit special Lagnas as distinct products, beginning with selected Bhava/Hora/Ghati Lagna traditions. Define sunrise/elapsed-time ownership, units, sidereal frame and lineage differences; provide typed results and transport provenance.	No first-class Bhava/Hora/Ghati Lagna implementation was found in the inspected engine/server tree. Ordinary chart Lagna, vargas and Pancha Pakshi clock inputs do not supply these definitions.	SOURCE_RESEARCH, P3

5. Gochar enrichment beyond the admitted snapshot

ID	Work still required	Current evidence and boundary	Status / priority
VED-017	Locally complete for reader-bound epoch/civil-date composition and bounded modern whole-sign windows: all seven subjects/blockers, all twelve ayanamsas, direct/retrograde re-entries, ordinary/baseline Vedha and optional once-derived natal BAV. Certain state, boundary bands and unresolved evidence share a complete partition.	The window standard, completed seven-stage plan and validation receipt record engine/facade/REST parity, 366-day bounds, external astronomy, resource budgets and shared-owner regressions. forecast.dated_events admits this window profile; existing snapshot choices are unchanged. Broader source profiles and modifiers remain VED-018/019.	LOCAL_COMPLETE; authorized engine/wiki source-publication package in section 49; release/deployment separate
VED-018	Evaluate additional Gochar corpora, references and nodal participation independently. Collate complete source profiles, baseline/indication tables, subjects, blockers and exemptions; resolve known edition/table discrepancies before exposing an executable choice.	Catalogue records profile.brihat_samhita_104, reference.alternatives, the conflicting Phaladeepika/Prasna Marga nodal treatments, nodes.blocker_participation and disputed tables. Seven classical bodies and natal Moon remain the only admitted computation profile.	SOURCE_RESEARCH, P3
VED-019	Evaluate separately sourced Gochar modifier/extension products: Counter-Vedha, source-specific sign-part activation, dignity/aspect and dasha context, ingress-Moon Murthi, and Nakshatra/chakra/Latta systems. Preserve relation types and competing conventions; do not collapse them into one strength score or generic Vedha switch.	Catalogue records vedha.counter_vedha, the two distinct activation treatments, strength-context records, murthi.ingress_moon and nakshatra.chakra_extensions. Each carries a non-admission limitation. Sarvatobhadra and 27/28-star geometry need their own source object, not an inferred extension of sign-based Vedha.	SOURCE_RESEARCH, P3

VED-018 and VED-019 are evidence programmes. Source-attested, disputed, research-required and outside-snapshot records remain visibly distinct. A catalogue entry is not authorization to compute it, and this register supplies no repaired disputed table. Remedies and interpretive recommendations remain outside the computational package.

6. Source, validation and documentation follow-through

ID	Work still required	Current evidence and boundary	Status / priority
VED-020	Preserve the remaining Pancha Pakshi source-admission queue: alternate solar/Padu/Sookshma doctrines, Bharana/Adhikara identities, vinadi routing, unresolved outcome cells, cross-witness composition, condition/scoring and electional search. Resolve each blocked identity/edition layer independently before runtime or REST promotion.	The current research standard, deferred products is the owning detailed queue. Stage 2J and the later research pilots are not blanket public outcome admission. Existing schedules, clocks, natal/Padu/EAT identity and other admitted capabilities must not be reclassified as absent.	SOURCE_RESEARCH, P3; retain the owning stage boundaries
VED-021	Strengthen independently traceable worked-case coverage for the specific extensions being admitted. Record source editions/pages, inputs, expected component results, boundary/missing-data cases and explicit tolerances. Use JHora/Kala differences to locate lineage disagreements rather than treating software agreement as classical authority or predictive accuracy.	Existing Vedic unit/server tests and standards already exist; the paused tracker saying validation is "None" is stale. The surface receipt records 397 focused passes and separate 193/72 regression runs for their actual scope; VED-005 adds 175 distinct admission/lifecycle/transport passes. VED-002 closes its selected source example, component/name/body/boundary and real-reader/HTTP coverage. These are not a validation study of every future rule.	EVIDENCE_REVIEW, P1 per selected package; VED-002 and all-seven-family VED-007 follow-through locally complete
VED-022	Reconcile remaining source headers and technical contract prose, then publish generated documentation when source publication is authorized. Require engine/public/REST closure receipts for future completion notices.	VED-001 already reconciled the Jaimini header to one/two-cycle support and named source limits. Muhurta policy/helper/personal-score/search prose is reconciled by VED-004/006; the initial June design is explicitly historical. VED-005 reconciles the selected family's REST standards/reference and generated pages. VED-002 reconciles Sayanadi calculation, chart/birth admission, legacy prose provenance and standards/reference/receipt. Generated moira.wiki is updated only through scripts/sync_git_wiki.py; source push and runtime release remain separate.	OPEN_INTEGRATION, P1 documentation work in each affected package; VED-002 and all-seven-family VED-007 follow-through locally complete

Authorized lunar-month and festival extensions, 6 October 2026

ID	Work still required	Current evidence and admission gate	Status / priority
VED-023	Source and implement exceptional Purnimanta month/fortnight mapping and Kshaya-year regional relabeling. Regional solar dates and era years require their own explicit profiles.	Lunar-month engine, Moira.lunar_month_at, curated exports and POST /v1/panchanga/lunar-month implement bounded Amanta/ordinary-Purnimanta context with solved conjunctions/ingresses, Adhika/Kshaya evidence, uncertainty brackets and explicit unsupported states. Source/policy/validation record. The PAC solar-ingress residual is reconciled: corrected IAU 2006 general precession; all 13 published ingresses pass the original 60-second gate.	BOUNDED_ADMISSION, P2; exceptional source rules remain open
VED-024	Admit finite festival/fasting rule catalogues by identified lineage: exact ritual windows, viddha, repeated/skipped tithi selection, intercalation eligibility, precedence and parana. Expose catalogue admission states and bounded typed evaluation through REST together.	Research and policy surface records institutional calendars, local corpus limits, explicit policy axes and test requirements. PAC dates do not constitute a complete executable rule specification; no universal festival catalogue is admitted.	SOURCE_RESEARCH, P2; research packet decision defer_for_source_completion

7. Optional expansion candidates

These are not promoted to required engine work by this reconciliation. Select a finite scope and evaluate sources before adding them to sections 2–6.

Candidate	What a scoping decision must settle
C-01: Vedic matching	Named Kuta/Ashtakoota/Dashakoota lineage, component calculations, regional exceptions and visible cancellation rules; chart compatibility ingredients do not establish an admitted matching product.
C-02: Prashna	A particular Vedic Prashna text/profile and required inputs; existing Western Horary evidence is a different doctrine.
C-03: KP astrology	Star/sub/sub-sub partitions, cuspal sub-lords, significators, time policies and named source/lineage. A Krishnamurti ayanamsa is not a KP engine.
C-04: Tajika/Varshaphal long-tail review	Inventory the finite source-owned cases missing from the already substantial Varshaphal engine before claiming any whole family absent. The May notes are a review lead, not a verified current defect list.
C-05: Research/corpus/filtering tools	Define licensed/provenance-bearing datasets, repeatable chart queries and component comparisons. This is infrastructure; it is not an additional Vedic calculation technique or evidence of predictive validity.

The user authorized lunar-month and festival investigation after VED-015. VED-023/024 now own that work; unadmitted regional calendars and ritual branches remain visible, rather than being folded into a claim of daily Panchanga completeness.

8. Work already present: do not reopen as missing

Existing product	Correct remaining boundary
Seventeen Varga divisions, Vedic dignities, Shadbala including Bhava Bala, BAV/SAV and reductions, Vimshottari/Ashtottari/Yogini	Existing families; do not reconstruct the pre-Phase-1 backlog. Deepening must name a specific rule or defect.
Yoga core, extended Jaimini, upagrahas, four chart-evaluated avastha systems plus optional Sayanadi	Existing engine and REST families. Selected yoga/Jaimini enrichment remains above; Sayanadi's bounded admission is closed locally.
Tara/Chandra Bala, personal Muhurta, Vimshopaka/Vargottama, Kakshya/Shodhya Pinda, Sade Sati	Existing calculations and REST. Their old "missing" notices are superseded.
Rao Chara second cycle	VED-001 locally closes strict cycle selection and canonical engine/REST receipts; complete-cycle and co-lord authority remains bounded in the owning standard.
D60 named deities	VED-003 locally closes eight-route transport, four selected-edition names, source-only signs and modern/classical-derived full-degree profiles. Each profile has explicit provenance and completed engine/facade/REST/strength validation. Direct classical D60 degree prescription and exact printed planetary pairs remain separately unclaimed; the absence of a direct prescription does not block the derived policy.
Sayanadi	VED-002 locally closes the selected BPHS source recipe, strict standalone/chart/node/birth inputs, inspectable results and curated Python/facade/REST surfaces. Legacy effect prose is explicitly uncertified and unevaluated.
Gochar snapshot, dated composition and bounded windows	Original snapshot package is source-published. VED-017 now closes the selected epoch/civil-date and modern window engine/facade/REST contract locally; VED-018/019 remain separate admissions.
Daily sunrise-owned Panchanga	VED-015 engine/public/REST is locally complete; its named policy, source and bounded verification are in the linked receipt. No website adoption or deployment is implied.
Unified Vedic exports	The 6 October package added 84 curated vedic names and sixteen root Varga names. Their absence is closed; this is not blanket admission of every standalone module helper.
Existing Muhurta, Vedic, Pancha Pakshi and Varshaphal standards/tests	They exist. Additional worked-case coverage and scope corrections are specific follow-through, not "no documentation" or "no tests."

9. Suggested package order and definition of done

Latest completed selection: VED-013 - source-defined Chara and Argala profiles. All stages of the seven-stage plan have a local execution receipt. Three Chara recipes, three bounded Argala assessments, legacy admission/receipt repairs and equally capable Python/facade/REST surfaces are implemented and validated. General strength comparison and the five named research candidates remain separate admission work. VED-013's authorized source-publication package is recorded in section 55. Section 56 records the subsequent locally validated result-integrity and HTTP error-classification repairs; section 57 records their authorized source-publication package. VED-014's subsequent research is recorded in section 58 and its named-profile engine/facade/REST completion in section 59. The source research checkpoint and implemented validation remain separate evidence. The next numbered open package is VED-016; its scope still requires selection rather than automatic continuation. VED-012's selected 21-entry package retains its section-52 source-publication checkpoint; BPHS 39.35 and broader additions remain research records. VED-011's four-area purpose package and VED-017's bounded Gochar windows retain their completed/source-published checkpoints. Release/deployment remain separate.

1. Close existing product omissions: VED-001/002/003/005 and their scoped VED-021/022 follow-through are complete and source-published. This first item's selected implementation scope is closed. Operational REST budgets remain a separate scope decision.
2. Muhurta composition: VED-004/006 are complete for the bounded sampled/JD-weekday profile. VED-007 is locally complete for all seven named families, including exact Sun/Moon transitions and sunrise-owned named-day composition. VED-008 is locally complete for the named source assessments and bounded day cells. VED-009 is locally complete for its six detectors and three source-scoped exceptions. VED-010 is source-published for its selected Lagna/Navamsa components; its optional Shadbala composition and subsequent admission defects have passed combined repair acceptance. VED-011's selected marriage and four-area purpose package are complete locally; further traditions/purposes and integration into generic scoring/search remain separate scope decisions.
3. Daily foundation: VED-015 is locally complete. Selected special Lagnas remain a separate selection. Kalachakra's named Santhanam/BPHS profile is locally complete under section 59. Its remaining traditions, deeper periods, gati taxonomy and eligibility rules retain explicit non-admission.
4. Deepen source-specific families: VED-012's selected 21-entry package and VED-013's selected six-profile package are locally complete. Full BPHS/PVR Chara, Narayana, Yogada and general Argala strength remain source-specific selections, with concrete gates in VED-013 discovery and section 54. Gochar VED-018/019 retain their research scopes. VED-017's bounded windows are locally complete. Pancha Pakshi retains its owning stage-by-stage queue.

Every admitted package must state the governing object, edition/lineage and ambiguity policy; input and output ownership; evaluated/omitted/unavailable distinctions; public export identity; explicit bounded request and typed response contracts; reader/resource lifecycle; direct-engine/HTTP parity; hostile and boundary inputs; source-owned fixtures; documentation and completion receipt. New output fields must preserve canonical engine truth rather than rebuilding it in the server. Numerical authority, transport fidelity and predictive claims are different validation questions.

For research-only items, completion is an evidence packet plus an explicit admission/defer/exclude decision. For calculation work, evidence precedes runtime implementation. For a public/REST gap, completion requires both surfaces and their contracts. A push is not a release, deployed endpoint or website tool.

10. Initial reconciliation evidence and maintenance

This is a source/contract reconciliation, not a new source-book collation or whole-engine numerical certification. Current code, model fields, public curation, registered routes, standards and tests were inspected. A kernel-free project-runtime smoke additionally verified:

• 12 versus 24 Chara periods and repeated sign/year/lord spans;
• absent REST cycle selection;
• absent Sayanadi chart inputs and all chart Sayanadi slots remaining None;
• absent root Sayanadi curation and absent REST birth inputs;
• absent D60 response deity and absent natal Muhurta request weights;
• absent claimed Muhurta dosha/Lagna-strength helpers;
• an unused natal parameter in the scorer's syntax tree;
• accepted string/boolean Panchanga numeric coercion.

Runtime: project Python 3.14.3, Moira 6.9.9, with MOIRA_NO_DOWNLOAD=1. The witness ran through .\.venv\Scripts\python.exe - with a stdin assertion script; it used no chart resource or network. The preceding 397/193/72 test receipts were not rerun for these documentation edits and are cited only for their original package scope. A filename search of the local book corpus found Raman PDFs, including a volume-2 file; the old statement that "no accessible scan exists" was not carried forward as a verified fact. Exact book/edition/verse identity still needs collation for the intended yoga cases.

The eight canonical documentation paths changed in this pass are this register; the three historical Vedic roadmaps/trackers; the Vedic entries in the broad frontiers register; section 6.2 of the August audit; wiki/Home.md; and the unified Vedic paragraph in wiki/02_standards/API_REFERENCE.md. Engine, REST implementation, numerical fixtures, test policy, dependencies, version, generated wiki, website and other worktrees are untouched. The initial working tree was clean.

Verification commands for documentation:

$env:MOIRA_NO_DOWNLOAD='1'
.\.venv\Scripts\python.exe scripts/check_doc_consistency.py
.\.venv\Scripts\python.exe scripts/sync_rest_api_reference.py --check
git diff --check

All three documentation commands passed. Direct project-runtime checks also passed for 22 unique ordered work IDs, five unique candidate IDs, 50 relative file/anchor links and five historical/superseded notices. An in-memory preview through the owning wiki generator passed with the new untracked register explicitly included; it verified flat-page and engine-source link rewriting without staging or writing generated files. Generated-wiki publication is a later authorized operation; this pass does not stage, commit, push or synchronize the mirror.

Keep the IDs stable. Close items with a linked source/engine/public/REST receipt, not a blanket "complete" label. Split a package only when its doctrinal object or independently reviewable delivery changes; avoid duplicating shared requirements or resurrecting a historical absence claim. Update the baseline and date whenever this register is reconciled again.

11. VED-015 local completion, 6 October 2026

The user selected VED-015 as the next implementation package. The daily standard owns the resulting source review, selected and deferred conventions, strict input/output contracts, actual commands and verification limits. Section 10 above is the earlier documentation-only pass; its "untouched" statement applies to that pass. Other numbered work packages and optional candidates retain their prior scope.

12. Personalized Muhurta completion and source publication, 6 October 2026

The user selected VED-004 plus VED-006 after the dated-Gochar snapshot push. The owning standard and receipt record the repaired existing profile, bounded search, public/REST admission and actual verification. Source collation for additional schools, Sunrise Vara search and exact transitions are not implied by local composition correctness. VED-005's remaining family audit and VED-007–011 retain separate boundaries. Earlier section-10 evidence describes the initial reconciliation, not this later code.

After validation the user authorized commit and push of this package. The containing engine/wiki source commits publish VED-004 and the bounded VED-006 repair, with the generated wiki published first. Release, deployment and website/Urania adoption remain separate operations.

13. VED-001/003 research checkpoint, 7 October 2026

The user selected bounded Chara cycles and D60 deity transport, requesting deep research first. The owning source packet records online primary-author research, the local book/article corpus, visually checked edition pages, fingerprints, source disagreements, current engine/HTTP witnesses and a finite implementation proposal. It identifies an article/book attribution error, partial second-cycle evidence, four D60 name discrepancies and a separate D60 placement convention. No alternative Chara school or new Varga arithmetic is admitted by a transport omission.

The baseline is 140 passing existing engine tests plus 10 selected Chara HTTP tests, with separate kernel-free probes. These are scoped regression receipts, not authority validation of every proposed rule. VED-001/003 remain open; the 21-open count is unchanged. Only this register and the new research packet are edited in this checkpoint. Runtime, validation policy/evidence baselines, generated wiki, source publication, release, deployment and website adoption are unchanged.

14. VED-001/003 implementation, 7 October 2026

The user approved the four-slice plan. Chara now validates bounded cycles and complete inputs and returns actual execution metadata through REST. All eight Varga placement routes preserve canonical nullable enrichment. Four deity names are corrected against the identified Santhanam printed list with explicit migration values.

Further source adjudication admits bphs_santhanam_sign as a typed sign-only product and selected Vimshopaka method. It does not supply continuous degrees; full-position consumers reject that selection and preserve harmonic defaults. The D60 standard records the missing passage/fixtures needed to close that subitem. The Chara standard distinguishes partial published repetition evidence from the retained co-lord convention.

The completion receipt records source boundaries, exact compatibility changes, focused tests, installed DE441 chart integration and documentation checks. VED-001 is locally complete; VED-003 remains bounded with full source placement open. Twenty numbered packages remain open; historical checkpoint counts above retain their dates. Commit/push, generated-wiki publication, release, deployment and website/Urania adoption were not authorized by this implementation request.

15. D60 full-position extension, 7 October 2026

The user authorized widening source review and exceeding the full-position acceptance checks. The new source packet adjudicates the dated PVR textbook/linear composition, author revision notices, the article's incorrect D9 arithmetic and unpaired published D60 coordinates. The named modern full profile is implemented through engine, facade, all seven named/chart request shapes and strength, with canonical positional source receipts, strict new-profile inputs and exhaustive half-degree boundary checks. The execution receipt records this extension separately from the earlier checkpoint above.

VED-003 remains BOUNDED_ADMISSION: named modern computation is locally complete; the classical passage and paired published full-D60 oracle are not established. This does not reset other package states or change the 24 stable IDs/20 open count. Source publication, release and product adoption remain separate states.

16. D60 bounded evidence decision, 7 October 2026

The completed adjudication reviews two identified BPHS editions/commentaries, named local treatments and further primary author publications. Retain the explicit dated Moira composition; admit no classical full-degree method or current JHora equivalence. Sharma's published sign example/table rows provide independent regression evidence. The single published arudha D1/full-D60 pair is compatible under nearest-minute rounding and incompatible under truncation; its rounding rule is unknown.

The finite research task is complete. The remaining classical/planetary-oracle claims are not established in the reviewed sources, with concrete evidence required to reopen them. This preserves VED-003's bounded admission and the 24 stable IDs/20 open count. No new runtime/API selection or source authority is introduced by the validation witnesses.

17. D60 broader computational validation, 7 October 2026

The completed broader receipt records 24,432 conditioned inputs compared with checksum-pinned PyJHora full-point outputs and compiled Maitreya returned signs/instrumented longitudes: all 720 segments, exact/adjacent boundaries, 20,000 seeded samples and 112 planetary inputs across sixteen date/frame charts. No disagreement exceeds the predeclared arithmetic budget; signs agree exactly. Independent rational evaluation agrees throughout. The 4,552-case offline fixture reproduces exactly; permanent engine/facade/REST/strength tests use external expected values. The source fragments are private validation inputs, not runtime dependencies or copied engine implementation.

Modern implementation and broader computational validation are complete locally. VED-003's remaining bounded state concerns historical authority claims, including an identified classical full-degree law and exact published planetary pairs. Neither is concealed as an unfinished engineering gate. Current PVR Jagannatha Hora equivalence and predictive effectiveness are not claimed. The 24 stable IDs/20 open count, other package states and existing source locators are preserved. Commit/push, release and website adoption are separate from this local outcome.

18. Classical-derived D60 reconciliation, 7 October 2026

The wider-library adjudication supersedes sections 14-17 wherever they imply that no classical-derived full-degree policy can be admitted. It establishes separately scoped BPHS D60 sign authority, Saravali general subdivision arithmetic, Jataka Parijata D12 fractional correspondence and Raman's explicit Navamsa degree notes. The D60 degree extension is Moira's disclosed derivation.

classical_derived_linear is implemented through the full engine/facade, all seven named/chart REST shapes, sign projection and D60-containing Vimshopaka groups. Canonical source locators and d60_degree_attribution make its derivation inspectable. Existing modern positional locators retain their exact identity. The completed validation receipt covers exact rational boundaries, primary commentarial D9 examples, unchanged frozen external witnesses, real DE441 charts and strict REST preflight. Source-profile groups also normalize tiny negative wrap inputs before composing the other divisions and D1 relationships.

The admitted classical-derived product and its engineering validation are complete locally. A directly prescribed classical D60 degree algorithm and exact source-published planetary pairs remain unclaimed historical frontiers. At this checkpoint the register retained 24 stable IDs and 20 open-or-bounded packages. The final closure below supersedes that count and VED-003's earlier bounded status. Publication, release and product adoption remain separate.

19. VED-003 final closure and source-publication package, 7 October 2026

The user approved closing the admitted D60 scope and publishing the completed engine/wiki package. VED-003 is LOCAL_COMPLETE. Its source sign policy, modern composed full positions, classical-derived full degrees, edition-owned deities, public/facade/REST transport and D60-containing strength consumers are implemented and validated. The execution receipt records 1,552 distinct passing tests and 24,432 conditioned corpus checks within the unchanged numerical budget. No engineering work remains for the admitted profiles.

The register now contains 24 stable IDs, five closed scoped packages (VED-001/003/004/006/015), and 19 open-or-bounded packages, including VED-017's independent window frontier. A directly prescribed classical D60 degree law and exact source-published planetary pairs are optional future research; neither is a closure gate or an additional promised work package. Historical checkpoint counts and decisions above remain dated evidence.

VED-001's already completed bounded Chara cycle work is included in the same coherent engine/wiki publication package. Generated wiki source is published first; the parent engine package then records that wiki gitlink. Package release, deployment and website/Urania adoption remain separate actions.

20. VED-005 local admission closure, 8 October 2026

The authorized audit and repair closes the selected 30 existing Vimshottari, alternate-dasha, Shadbala, dignity, Sade Sati and Vedic-profile contracts. VED-005 is LOCAL_COMPLETE at that scope. The admission standard and validation receipt record the original failures, reader repair, strict policy/input admission, alternate chart-frame correction, applied receipts, 175 distinct latest passing tests and actual startup checks. Previously coerced inputs, unknown classical keys/lords and unsupported house identities intentionally fail validation. Valid partial dignity maps and selected mixed component frames retain their existing meaning.

The register now contains 24 stable IDs, six closed scoped packages (VED-001/003/004/005/006/015), and 18 open-or-bounded packages. Section 9's first package now leaves VED-002 Sayanadi evidence/admission as the remaining technique, with per-package VED-021/022 follow-through. Jaimini's already corrected header is reconciled here rather than reopened as a defect.

No new classical calculation or full-family authority claim is introduced. Range, execution and response budgets for the existing Sade Sati/deep Vimshottari services remain a separately scoped operational frontier. Local closure does not imply source push, release, deployment or website/Urania adoption. Historical counts above retain their dated checkpoint meaning.

The user authorized source publication on 8 October: commit/push the generated wiki first, then the parent engine's scoped code, tests, canonical documents and wiki gitlink. VED-002 research begins after that checkpoint; it does not change the VED-005 validation or admit a new classical calculation.

21. VED-005 source publication, 8 October 2026

The authorized publication completed in the requested order:

• Generated wiki 293979c942b4083f59569f58a0e7b0d697bdbe39 was committed and pushed to origin/master first.
• Parent engine 77d34158cef7c936fdf6613f9be18ba5e8fad5aa was committed and pushed to origin/main second, including the matching wiki gitlink.
• Both checkouts were clean and at zero divergence after publication; exact remote branch SHAs and generated-wiki synchronization were verified.

The 175 distinct latest passing tests remain the original selected VED-005 validation scope. This publication pass reconciled status prose and performed documentation/whitespace checks; it did not alter the validated calculations, test policy, version 6.9.9, release or deployment state. The subsequent Sayanadi research documents are a new local package, not part of those published SHAs.

22. VED-002 source review and execution plan, 8 October 2026

The user selected Sayanadi after VED-005 publication. The source research collates local Santhanam BPHS, an identified Sharma edition, Rao's textbook and a competing Sanketanidhi/Hora Ratnam degree reading. Visual page review establishes the selected BPHS within-sign Navamsa-ordinal profile, sunrise-owned ghati ordinal, five-group name table and all nine body constants. The printed Sun arithmetic is consistent; the existing fixture's Sa=1 is incorrect and masks a different first substate remainder.

The unmodified engine's 34 selected tests pass, while an independent exact rational classification audit demonstrates 1,015 disagreements in 14,580 boundary cases. Missing Moon/body fallbacks, hostile input admission and absent chart/public/REST composition also require repair. The implementation plan defines four finite slices and their source, boundary, strict-input, time, reader, public/REST and documentation gates. Historical effect prose and competing degree-based variants retain their own source boundaries.

VED-002 remains open for engine and integration work. Research completion does not close its implementation or VED-021/022 follow-through. The register retains 24 stable IDs, six closed scoped packages and 18 open-or-bounded packages. No VED-002 runtime, native substrate, dependency, golden baseline, website or deployed endpoint changed in this source-audit checkpoint.

23. VED-002 local implementation closure, 8 October 2026

The user approved execution of the four-slice Sayanadi plan. The admission standard and validation receipt close the selected BPHS within-sign Navamsa-ordinal calculation, strict name/ghati/body contracts, optional chart/node evaluation, reader-bound sunrise/birth composition and owning Python/facade/REST surfaces. The corrected source example, all source constants/tokens, 14,580 independent rational boundary cases with zero disagreements, analytical uncertainties, four real date/frame cases, polar absence and both startup-reader paths are verified. The distinct latest test total is 278, with no failures/errors/skips.

VED-002 is LOCAL_COMPLETE; its scoped VED-021/022 validation/documentation follow-through is complete. The current register has 24 stable IDs, seven closed scoped packages (VED-001/002/003/004/005/006/015), and 17 open-or-bounded packages. Section 9's first item has no remaining selected implementation work. Historical sections 20-22 retain their earlier checkpoint counts and are superseded by this execution receipt.

The supplied-input and birth-clock contracts preserve omitted, evaluated and unavailable distinctions. Legacy effect summaries carry explicit uncertified, unevaluated provenance. Competing degree-lineage admission and a full textual catalogue audit are independent source scopes, not hidden incomplete work in this admitted numeric profile. Astronomical/native substrate, version, dependencies, test policy and unrelated work are unchanged. Canonical and generated documentation are reconciled locally; this package is uncommitted and unpushed. Release, deployed REST and website/Urania adoption are separate.

24. VED-002 authorized source publication, 8 October 2026

The user authorized committing and pushing the completed Sayanadi package before proceeding with VED-007. The containing source commits publish the generated wiki first, then the parent engine's code, tests, canonical documentation and matching wiki gitlink. Section 23's uncommitted statement describes its earlier local-completion checkpoint. The 278-test receipt and 14,580-case rational audit retain their recorded verification scope.

This is source publication at version 6.9.9, with no release or deployment. VED-007 begins after this checkpoint as a separate named-Muhurta package.

25. VED-007 two-window admission, 8 October 2026

The preceding Sayanadi publication is verified at engine acf5851c3b2369a9dd5630921437dc31c7064b4d and generated wiki 447768124157bc8df4d8be9347506da2d831b98d, with matching remote SHAs and clean checkouts at that publication checkpoint.

The authorized VED-007 package then admitted Abhijit and Brahma intervals after reviewing the Avasthi Chintamani scan, Arunadatta/Hemadri commentary, Raman transcription and local-library witnesses. Its canonical engine owns source and compatibility policies, supplied/solved solar anchors, root/endpoint bounds, per-window availability, civil sunrise-date ownership and separate weekday eligibility. Seven exports, two facade methods and two strict REST routes preserve that same result.

The 224-test receipt records independent rational formulas, source discrepancy handling, boundary/hostile/partial cases, real DE441 polar/DST events, unchanged PAC sunrise component tolerance and both serving-reader startup paths. Source-formula, component-astronomy and HTTP evidence are stated separately. VED-021/022 follow-through is complete at this scope.

The two-window integration is complete locally; VED-007 remains a bounded package because its individually named additions still require admission. There remain seven closed packages and seventeen open-or-bounded packages. The new code and documentation are uncommitted/unpushed. Source publication, release, deployed REST and website/Urania adoption remain separate.

26. VED-007 authorized source publication, 8 October 2026

The user authorized committing and pushing the completed Abhijit/Brahma package. The containing source commits publish the five generated wiki pages first, then the parent engine's implementation, tests, canonical documentation and matching wiki gitlink. Section 25's uncommitted statement records the earlier local-completion checkpoint. The 224-test receipt retains its actual source, numerical, real-resource and transport verification scope. Publication changes status and clarifies remaining scope; it does not alter the validated arithmetic or acceptance thresholds.

Godhuli, Vijaya, Amrita, Ravi Yoga and Sarvarthasiddhi are still outstanding parts of VED-007. Their rules have not yet completed source review, implementation and validation. SOURCE_RESEARCH is not evidence that classical sources are unavailable. VED-007 remains partially complete as a whole, with its two-window engine/public/REST contract complete. Version remains 6.9.9; source publication is separate from release and deployment.

27. VED-007 five-name local completion, 8 October 2026

The user authorized finishing the five outstanding names. The earlier sections 25–26 describe the historical two-window checkpoint; their research status is superseded for Godhuli, Vijaya, Amrita, Ravi Yoga and Sarvarthasiddhi.

• Primary source collation now supplies explicit Godhuli width/visibility restrictions, Vijaya's eleventh daylight part, both separately named Amrita tables, Ravi's inclusive Sun/Moon count and Sarvarthasiddhi's weekday table. The source packet records editions/pages/hashes, disagreements and the unidentified print edition of the digital Muhurta Sadhana witness.
• moira.special_muhurta supplies all five additions and a dated seven-name composition retaining the existing Abhijit/Brahma object. Eleven owning exports, three facade methods and three typed REST routes complete access.
• Numerical root bands, source eligibility, empty matches, missing astronomy and unevaluated activity suitability remain distinct. Sun and Moon star transitions both bound Ravi; weekdays belong to the requested sunrise day.
• Final combined acceptance: 383 passed, no failures/errors/skips; 33 DE441 resource uses, all run. Complete source-table/pair coverage, independent rational and analytic cases, real DST/polar/reader/HTTP cases, and existing-family regression are documented in the linked receipt.
• Fourteen Python files pass 3.10 grammar checks with no new Ruff findings. REST inventory is 489 paths/operations. Canonical and generated docs are reconciled; no new runtime dependency or substrate change was introduced.

VED-007 is LOCAL_COMPLETE for its seven source-selected timing/presence families. The alternative Amrita readings are supported, not conflated. Universal activity elections, other Godhuli seasonal/optical readings and other names called Amrita are not silently claimed; VED-008–011 retain their own rule/purpose scope. Eight of the 24 packages are complete, sixteen remain open/bounded. This paragraph records the local checkpoint; section 28 records the subsequent source-publication authorization. No deployment is implied.

Owning evidence: source packet, standard, and execution receipt.

28. VED-007 authorized complete source publication, 8 October 2026

The user authorized committing and pushing the completed five-name extension. The containing engine/wiki package publishes all seven named families across engine, curated Python/facade and REST surfaces. The generated wiki is published first; the parent then records its exact gitlink. The implementation acceptance remains 383 passed, zero failures/errors/skips, with 33 DE441 resource uses, all run. Publication changes status prose, not computation.

The source-selected completion boundary remains explicit: Amrita Siddhi and Kalaprakasika Amirtha are different profiles; Godhuli's selected timing and weekday exceptions are inspectable; numerical uncertainty is retained. Eight of 24 stable packages are complete and sixteen remain open/bounded. Source publication does not imply a package release, deployed endpoint or website.

Recommended next: VED-008 — Panchanga Shuddhi. Establish named source rules for Tara-cycle/paryaya distinctions, Panchaka Rahita and finer Yoga/Karana eligibility, then provide canonical engine results and equally strict REST transport. Existing Tara/Chandra Bala must remain intact. This recommendation does not start that separate implementation package.

29. VED-008 research and proposed admission, 8 October 2026

The user authorized research following the completed VED-007 publication. The source research packet collates online primary witnesses and local corpus scans, including the 1946 Sastri/Bhat Brihat Samhita, image-only chapter 100, and P. S. Sastri's textbook filed under the Raman author folder. Printed pages and scan hashes control attribution; OCR absence is not source absence.

The proposed finite package covers Panchaka Rahita, separately selected MC/PS Tara-cycle rules, the nine adverse Nitya Yogas with partial periods, and an eleven-Karana catalogue plus Bhadra components. It states clock derivations, exception conflicts, strict engine/REST contracts and concrete acceptance cases. The private arithmetic check proves the KP offset formula and direct remainder table agree in all 68,040 bounded inputs and checks two printed examples; it is not astronomical or predictive validation.

Research is complete for these admission decisions. The inconsistent MC Hindi thirds interpretation and Raman initial-ghati example have specific recorded deferral conditions; they do not prevent the recommended named profiles. Universal Bhadra exception precedence is unclaimed. Runtime implementation, REST additions, scoring/search integration and publication have not started. Eight stable packages remain complete and sixteen open/bounded.

30. VED-008 local implementation closure, 9 October 2026

The user approved implementing the named profiles together while preserving existing Tara/Chandra behavior. The complete selected package now includes Panchaka arithmetic/current Lagna, separate MC and PS Tara-cycle readings, nine-Yoga timed restrictions, eleven Karana activity entries, and independent Bhadra residence/day-night/mouth/tail claims. Derived clocks and simultaneous restrictions/exceptions are visible; no universal favorable verdict is added.

Sixteen owning exports, three facade methods and catalogue/direct/day REST routes preserve the engine result. Sunrise-owned cells use full solved parent events, changing Lagna, coherent true ayanamsa, explicit root bands, bounded work and typed polar/missing-component states. The existing score/search and Tara/Chandra contracts are unchanged.

Acceptance totals 622 passed, zero failures/errors/skips, with 92 DE441 resource uses, all run: 608 new/compatibility unit/server cases and 12 new real-kernel integration cases, plus two final analytic solar-admission cases. Source tables, all 68,040 Panchaka arithmetic inputs, both Tara variants, rational temporal windows, strict REST, DST/polar and configured/discovered reader paths are covered in the receipt.

Nine of 24 stable packages are complete; fifteen remain open/bounded. Implementation, source research and canonical/generated documentation are local and uncommitted/unpushed. Version 6.9.9 is unchanged; no release, deployment or website/Urania work is included. VED-009 cancellation rules, VED-010 strength overlays and VED-011 complete purpose elections retain their own scope. Particular source disagreements remain named exclusions rather than silently merged rules.

31. VED-008 authorized source publication and next package, 9 October 2026

The user authorized committing and pushing the complete VED-008 package, including the source research, engine/public/facade implementation, three strict REST routes, independent fixtures, tests and canonical/generated docs. Publish the generated wiki first and commit its exact gitlink with the engine. The validation receipt retains the 622-test acceptance and source/clock/coverage boundaries. Nine stable packages are complete; fifteen remain open/bounded. Source publication does not imply a version release, deployment or website/Urania adoption.

Recommended next: VED-009 — Muhurta dosha and Parihara. First collate the source-specific Vishanadi/Tyajya, Yamaghanta and finer Gandanta detection rules, clocks, exceptions and boundary ownership. Then select a finite initial profile set and separately named cancellation rules. Every neutralization must retain the detected condition, the source rule and the evidence that its prerequisites are met; missing evidence must remain unknown. Existing Shuddhi restrictions/exception overlaps must not silently acquire universal precedence.

The subsequent implementation should reuse the bounded dated composition, carry source-separated detected/neutralized/unavailable/excluded states, and arrive with canonical public/facade and equally strict REST support. Independent source fixtures, exact boundary cases and real-reader/HTTP parity are its acceptance gates. VED-010 then adds selected Lagna/strength/Navamsha composition; VED-011 builds complete purpose profiles on those evaluated components. This sequence is a recommendation, not authorization to start those packages.

32. VED-009 source-selected local completion, 9 October 2026

The user selected the VED-009 package after the published release-hardening repair. The source packet collates inspected MC Avasthi, Kalaprakasika and local Santhanam BPHS scans, records the Rohini worked clock example and preserves differing Mula, Purva Bhadrapada and tithi-Gandanta readings as named policies.

The canonical dosha/Parihara owner implements stellar Vishanadi, weekday/star and daytime Yamaghanta, and temporal nakshatra/tithi/Lagna Gandanta. Three opt-in exceptions retain raw witnesses and nullable prerequisite evidence. Cross-source exceptions are excluded; missing evidence cannot manufacture an affirmative cancellation. Dated composition solves full parents, retains fixed-width carryover and merges all overlapping numerical bands, including edges whose estimates fall just outside the sunrise day.

Fifteen owning exports, three facade methods and catalogue/direct/day REST routes provide strict, lossless access. 153 new tests pass, including independent source/rational and analytic cases, six real-date component comparisons, polar/DST/reader cases and both configured/discovered HTTP startups. Existing named/Shuddhi regression tests also pass. The validation receipt records the separate release-gate and documentation checks without treating internal numerical agreement as an external astrological oracle.

Final local acceptance totals 1,999 passed, one existing unrelated void-of-course skip, zero failures/errors. All 55 Release Hardening test files and all six generated/release documentation checks pass; DE441 records 231 successful resource uses across the feature and release runs. The REST inventory has 495 operations. Exact public snapshots and class-docstring governance include this package, without weakening their assertions.

VED-009 is LOCAL_COMPLETE for this finite six-detector/three-exception package. Ten of 24 stable packages are complete; fourteen remain open/bounded. Scoped VED-021/022 follow-through is complete. The full twenty-one-dosha catalogue, other Tyajya types and angular/Abhukta variants remain explicitly listed source extensions. These exclusions do not imply unavailable sources.

Recommended next: VED-010, selected Lagna/strength/Navamsa composition and its dependent exception prerequisites, followed by VED-011 purpose profiles. Neither package is started here. Current VED-009 work is local, uncommitted/unpushed at version 6.9.9; release, deployment and website/Urania adoption remain separate.

33. VED-009 authorized source publication, 9 October 2026

The user authorized committing and pushing the completed VED-009 package, then proceeding with VED-010. Publish the generated wiki first, followed by the engine source, tests, canonical documentation, publication manifest and exact wiki gitlink. Section 32's uncommitted statement records its earlier local checkpoint. The acceptance remains 1,999 passed, one existing unrelated void-of-course skip and zero failures/errors; publication does not change arithmetic or validation thresholds. Version 6.9.9 remains unchanged.

VED-010 is now authorized as the next engine/REST package. Begin with source collation for Lagna, planetary-strength and Navamsa-shuddhi composition and its named purpose/exception prerequisites; reuse existing Varga/Shadbala owners without hiding school choices or missing inputs. VED-011 complete purpose elections remain separate. Source publication does not imply a package release, deployment or website/Urania adoption.

34. VED-009 publication and VED-010 selected composition

VED-009 was committed and pushed wiki-first: generated wiki ed68937510e6ce162da82420b4f42415c0480161, then engine 571788a1fe705b096d5a69911f30be583c7c45d7. Remote branch tips and the parent gitlink were verified. Hosted Release Hardening and Acceptance Matrix completed successfully for that engine SHA.

The user then authorized VED-010. Source inspection established finite general and marriage Lagna profiles, named Navamsa variants, fractional-aspect support, MC87/92's placement score and scoped MC88 exceptions. Shadbala is separate context, not a blanket override. Source availability is established; broader remedies and VED-011 complete purpose elections remain explicit extensions.

The shared Varga D9 partition and Venus-to-Moon natural relation were corrected against BPHS6.12 and3.55. Their dependent regression evidence is part of this package. See the linked source record, standard and validation receipt for final acceptance status. VED-010 is included in the authorized wiki-first source-publication package.

VED-010 acceptance is complete locally: 363 new feature tests, 881 affected existing Vedic regressions, and all four release-hardening groups. The deduplicated final outcome is 2,856 passed, one existing no-aspect void-of-course skip, zero unresolved failures/errors. All 18 changed Python files pass Python3.10 grammar; full scoped Ruff, exact API/docstring guards, all six release-facing artifact checks and whitespace checks pass. The REST inventory now has 498 operations; the new unit/REST files are added to CI.

Eleven of 24 stable packages are locally complete; thirteen remain open/bounded. Next at this checkpoint: VED-011, selected complete purpose elections; section 35 subsequently puts the identified repairs first. This next package is not started by VED-010 closure. The user subsequently authorized VED-010 commit/push, followed by a deep adversarial review of the current Vedic system. This publication includes the generated wiki first, then the engine source, tests, canonical documents, publication manifest and exact gitlink. The review is a separate assessment of the resulting snapshot.

35. VED-010 publication and adversarial repair priority

VED-010 was published wiki-first at 9d3591367af6311bbec5f47afeab7918939b5724, then engine b578ec89b90f709c8e64d5d9405ca06d29fcc69c. Hosted Release Hardening and Acceptance Matrix succeeded. The separate adversarial review identified eleven findings: four P1 and seven P2, comprising ten runtime/contract defects and one stale validation fixture. Their root causes predate VED-010.

The completed broader regression run was 4,387 passed, one failed, zero errors/skips, across 112 files, with all 406 DE441 resource receipts run. The failure reproduces alone and is attributable to the Einstein alternate- Dasha fixture retaining balances derived with the superseded precession scalar. Independent source/boundary/HTTP probes establish the other findings, including a valid dated Lagna request failing only when Shadbala is included. The broader run is not green; the narrower release gates do not override it.

The user requested a plan covering all eleven repairs. It groups full-parent Dasha timing and fixture reconciliation, Varga boundaries, source-defined Shadbala components, war allocation/evidence, and reader/scan robustness into five packages with individual and combined acceptance. VA-01 through VA-11 remain open. This is a planning checkpoint, with no corrective implementation or new publication. VED-011 follows their closure.

36. All eleven adversarial repairs closed locally — 9 October 2026

The user approved the five-package plan and implementation. VA-01 through VA-11 are now closed across engine, curated/facade and REST surfaces. Dasha subperiods use full parents before clipping; Varga boundaries use one exact partition; source-selected strength components use explicit geometry; one canonical ledger governs all war adjustments; borrowed readers and Sade Sati work limits are enforced. The shared temporary-friendship correction and the geocentric REST strength frame are documented dependent repairs.

The completion/source receipt records 6,276 distinct passing tests, one existing unrelated VOC skip, zero unresolved failures/errors, across 173 files. Every original 112-file adversarial selection was rerun, plus new/affected consumers and all 62 current release-hardening files. Five initial migration failures were resolved with exact comparisons retained; the 23-file reconciliation passed 1,215 tests. All 835 DE441 resource uses ran. Independent 65,880-case Varga and 4,320-case D60 consistency corpora remain green. Source disagreements, retained conventions and sampling limits remain visible in the policy receipts.

This closes the repair checkpoint, not the remaining Vedic roadmap. Eleven of 24 VED packages remain complete and thirteen open/bounded; VA identifiers are repairs, not new VED packages. Next: VED-011, selected complete purpose elections. It is not started here. All changes remain local, uncommitted and unpushed at 6.9.9. Publication, release and deployment remain separate actions.

37. Second adversarial pass: five repairs closed locally — 9 October 2026

The second review identified five additional composition/admission defects. The approved repair now unifies dated Shadbala's UTC weekday evidence, canonicalizes Lagna/Bhava positions against their accepted strength context, recomputes house-independent strength components during receipt validation, checks Dasha full-parent/child and year provenance across engine/REST, and uses one half-open Sade Sati longitude normalization. Existing partial-child REST requests remain admitted through an explicit validator mode; generated trees continue to require complete coverage. Declared contradictory evidence rejects in both modes.

The updated validation receipt records 6,342 distinct passing tests, one existing unrelated VOC skip and zero unresolved failures/errors, across 175 files. The 121-file Vedic run and all 64 release-hardening files completed; a final 17-file reconciliation passed 1,104 tests, including all three compatibility regressions identified by those runs. All 814 DE441 resource receipts in the three runs executed. The 66 new regression cases are wired into CI. Original adversarial probes, 250 generated Dasha trees / 166,118 nodes, all six documentation checks, grammar, scoped Ruff and whitespace checks also pass.

This closes VA2-01 through VA2-05. The VED package count and remaining policy boundaries are unchanged; VED-011 remains the next separately authorized package. The complete repair work remains local and uncommitted/unpushed.

38. Combined repair publication package — 10 October 2026

The user authorized committing and pushing all accumulated repairs: VA-01 through VA-11 and VA2-01 through VA2-05, with their regression tests, CI selection, owning standards and validation receipt. Publication order is the generated wiki first, then the engine and its exact wiki gitlink. Sections 35–37 preserve the review and local-completion states at their earlier checkpoints; their uncommitted/unpushed statements are historical.

The combined acceptance remains 6,342 distinct passing tests, one existing unrelated VOC skip and zero unresolved failures/errors, with the resource execution and source-policy limits recorded in the validation receipt. The roadmap now consistently identifies VED-011 as the next package: select finite purpose profiles, collate their sources and disagreements, then admit engine and REST rules with explicit evaluated/excluded/unavailable evidence. Marriage variants, construction/Vastu, travel and business are candidate scopes, not a promise of universal activity coverage. Research and implementation of VED-011 have not begun in this publication package.

Version remains 6.9.9. Source publication, release and deployment remain distinct; no tag, package release, runtime deployment or website/Urania adoption is included.

39. VED-011 marriage-election research — 10 October 2026

The user authorized deep source research, starting with marriage elections, both online and in C:\dev\ASTROLOGY-BOOKS-DATABASE. The research packet records twelve fingerprinted PDFs, targeted visual collation, online primary and author sources, source disagreements, the current engine/REST boundary and concrete acceptance criteria. Image-only local sources were inspected as scans; failed text extraction was not treated as source absence.

The shared eleven-star wedding list differs from the legacy guidance, whose tithi sets also overlap. Source-defined 28-star vedha, Latta, marriage-specific Bana, additional Lagna restrictions and personal completeness need explicit admission. Jupiter/Venus availability has a classical computational source: Surya Siddhanta IX uses directional heliacal events and degrees of time, while MC prescribes directional waiting periods. Existing longitude-orb combustion is a separate convention and cannot silently substitute for it.

The proposed first scope is an MC Avasthi 2004 ordinary marriage astronomical timing profile, with separately declared personal and Godhuli layers. The implementation plan must enumerate every required rule, named interpretation and explicit exclusion before admitting complete selected-profile engine and REST results. Modern calendar dates are only partial comparison evidence; their omitted rules cannot validate the full composition.

This is a research checkpoint, not runtime completion. The eleven-complete / thirteen-open feature count is unchanged. No runtime, tests, protected validation policy, dependency or version changes are included, and there is no commit, push, release, deployment or website adoption in this pass.

40. VED-011 selected marriage admission — 10 October 2026

The approved plan is implemented and locally accepted. The source standard and validation receipt own the exact MC Avasthi ordinary, personal and VV/MC Godhuli scope. Forty-one curated symbols, four facade methods and four REST operations share the same raw evidence, named remedies, source coverage and bounded numerical results.

All 227 focused checks pass, including the complete six-hour DE441 HTTP partition and the high-latitude unavailable-evidence request. Source arithmetic, independent ERFA geometry, native record access, boundary/resource tests and engine/HTTP projection remain distinct evidence classes. The broader latest result set has 3,901 distinct passes and three unrelated disclosed skips. Operational costs, actual uncertainty bands, failed-attempt repairs and conditional arithmetic assumptions are recorded rather than hidden.

This closes the selected marriage implementation. Construction/Vastu, travel, business and other source traditions remain open VED-011 selections, so the 11-delivered / 13-open-or-bounded feature-package count is unchanged. The changes are local and uncommitted; publication and release are separate.

41. VED-011 marriage source publication and remaining scope — 10 October 2026

The user authorized committing and pushing the complete selected marriage package. Publish the generated wiki first, then the engine, tests, canonical documentation, publication metadata and exact wiki gitlink. Section 40's uncommitted statement records the earlier implementation checkpoint. Publication starts from engine e4e9a72 and wiki 9edb51f, preserving the separately published small-body and visibility-documentation changes since the validation baseline. All 41 marriage source-manifest fingerprints and seven aggregate XML receipt fingerprints still match at this checkpoint. The focused publication rerun passes 225 tests with eleven executed DE441 resource uses and no skips; the two full HTTP cases retain their unchanged successful acceptance receipts.

The remaining VED-011 implementation scope is:

Scope	Work still required
Construction/Vastu	Select and collate finite foundation/construction and house-entry profiles separately; implement their source-specific calendar, Lagna, direction and exception rules where established. Existing activity guidance is not a complete dated election.
Travel	Select finite journey profiles and establish their direction/purpose rules, exceptions and boundaries before admitting complete engine and REST assessment/search. Existing travel guidance is not sufficient.
Business	Select the specific activities and source profiles to support, then implement and validate their complete elections. A generic business label does not establish one universal rule set.
Additional marriage traditions	Admit complete alternative source profiles from the comparison ledger, retaining disagreements, their own finite remedy targets and profile-specific validation. The current ordinary profile remains MC Avasthi 2004 with separately selected personal and VV/MC Godhuli composition.

Each selected extension requires the same engine/public/facade/REST contract, explicit evaluated/excluded/unavailable evidence, source fixtures, boundary tests and real-reader/HTTP validation. The table does not authorize a new implementation or assert that sources are unavailable.

At this source-publication checkpoint, the tested full marriage HTTP calculations took about 25–28 minutes; section 42 records their subsequent optimization. Performance improvements must preserve the source predicates, uncertainty bands and completeness certificates; background-job orchestration belongs to a separately selected service scope. Whole-ceremony-duration coverage is an optional new marriage contract; the implemented search elects ritual-anchor instants. Compatibility matching, family/preparatory ceremonies and the other explicit plan exclusions are separate scope choices, not unfinished required rules in the admitted profile.

The selected marriage implementation is complete. The overall VED-011 entry remains open for the other purpose/profile selections above. Version 6.9.9 is unchanged; source publication does not imply a tag, package release, deployment or website/Urania adoption.

42. VED-011 marriage performance optimization — 10 October 2026

The marriage package is source-published at engine 03dc7e0 with wiki 2ce8edc. The user then authorized performance optimization while preserving accuracy and validation. The performance receipt records the locally completed change: one private native interval-Chebyshev primitive, with its exact Python reference retained and strict floating-point compilation confined to that primitive's translation unit. Source rules, search tolerances, supporting domains, resource limits, public contracts and work counters are unchanged.

The six-hour HTTP window calculation improves from 28.4 to 8.3 minutes; the high-latitude HTTP instant improves from 25.2 to 7.4 minutes. Both are approximately 3.4x faster in the measured runs. Each complete calculation response matches its original after accounting only for the new native backend fingerprint; the personal direct worked case is byte-identical. The paired same-process crossing benchmarks corroborate 2.6–3.9x speedups with identical roots, certificates and counters.

Acceptance has 345 distinct passing tests, six unrelated disclosed skips, both full real-kernel HTTP cases executed, bit-exact primitive/serving-record parity and independent rational first/second derivative checks. Isolated native checks also verify rejection of unsupported host arithmetic modes. The optimization is complete locally and remains uncommitted/unpushed; version 6.9.9 is unchanged. Full synchronous requests still take minutes. Other VED-011 purposes, additional marriage traditions, whole-ceremony extensions and service orchestration retain their separately selected scopes.

43. VED-011 exact computation reuse — 10 October 2026

The user authorized a second marriage performance pass, including reusable infrastructure for other engine computations. The reuse standard and validation receipt record the completed local implementation. A private standard-library helper provides bounded, thread-owned reuse of immutable results under exact byte keys. It is independent of marriage and is tested with a standalone NAIF TT/TDB consumer. Another feature must bind its own stable sources/policy and validate its admission; no shared cross-request result service is introduced.

Marriage uses five bounded memos for TT/TDB differentials, record arguments, frame polynomials, frame angles and trigonometric pairs. Serving-reader/source admission and public work accounting remain outside them. Original operation order, numerical bounds, tolerances, resource limits and engine/REST semantics are unchanged. No new native source or build change was required.

The six-hour HTTP window calculation improves from 8.3 to 6.8 minutes (497.29 to 409.20 seconds); the high-latitude assessment improves from 7.4 to 5.8 minutes (442.40 to 347.89 seconds). These observed reductions are 17.7 and 21.4 percent. Both complete responses match their predecessors with no response-field normalization; the personal direct worked case is byte-identical. The two requests record 371,422 and 313,401 successful helper reuses while retaining the same source evidence, certificates, unavailable states and counters.

Acceptance has 374 distinct passing tests, six unrelated optional skips, both full DE441 HTTP cases executed and no unresolved failures. Exact key, dependency isolation, eviction, exception, immutability, thread ownership, request lifetime and budget behavior are tested. The paired benchmark provides separate timing/equality evidence; historical whole-request timings do not establish a universal latency guarantee.

Both performance passes remain local and uncommitted/unpushed, on version 6.9.9. Full calculations still take minutes. Other VED-011 purposes, further source profiles and service orchestration remain separately selected scopes.

44. Both marriage performance passes — source publication checkpoint, 10 October 2026

The user authorized committing and pushing both completed performance passes together. This checkpoint supersedes the local/uncommitted publication status recorded at acceptance in sections 42 and 43 and their linked receipts. Those acceptance measurements and evidence remain historical records of the tested implementation.

Publication starts from engine 03dc7e0 and wiki 2ce8edc. All twelve distinct computational, native-build, benchmark and test source fingerprints across the two passes match their final validation manifests, using the reuse pass's astronomy module fingerprint where it supersedes the first pass. The native binary, five final reuse receipts and both aggregate test XML inputs also match their recorded fingerprints. Acceptance remains 374 distinct passes and six disclosed unrelated optional skips, including both successful full DE441 HTTP cases. No computational change is introduced by publication.

The publication sequence is generated wiki first, then the engine with that exact wiki reference. Release-facing documentation, release identity, Hellenistic and REST inventories, website documentation metadata, wiki synchronization and changed-line whitespace are checked before committing. Version 6.9.9 and the separately selected remaining VED-011 scopes are unchanged. This checkpoint covers source publication; it does not authorize a tag, package release or deployment.

45. Four remaining VED-011 areas — research and plan, 11 October 2026

The user authorized research into construction/Vastu, travel, business and additional marriage traditions, followed by an implementation plan. The research packet records twelve fingerprinted PDFs, six from the local book database, previous online witnesses, two newly acquired complete MC scans and a digital Sanskrit witness. Selected printed pages were visually checked against extracted text.

The old 207-page MC04 research scan ends partway through travel; its missing final chapters were recovered in complete scans, including a verified 1948 Maniprabha edition. Foundation and entry use different 28/27-star chakras. First entry, return and qualified extraordinary entry retain different scopes. Buying and selling have different preferred stars, travel requires explicit direction/anchor/exception semantics, and VV ordinary marriage has its own calendar and non-cancellable Jamitra requirements. Printed KP contradictions are retained as edition-specific collation decisions rather than guessed away.

The plan contains seven dependency packages: source closure, minimal shared evidence computation, housing, business, travel, additional marriage profiles and integrated acceptance. Each vertical includes strict engine/facade/REST contracts, source and boundary fixtures, actual real-resource instant/window acceptance and preservation of the published marriage/performance baseline. Remaining interpretation questions have explicit closure criteria.

Research arithmetic checks validate the transcribed finite tables and worked examples; they do not certify new engine behavior. This pass changes canonical roadmap documentation and its generated wiki representation only. Runtime, protected numerical/validation files, version, release and deployment are unchanged. Implementation remains proposed; VED-011 remains open and the eleven-delivered/thirteen-open stable feature counts are unchanged.

46. Four-area VED-011 implementation and closure — 11 October 2026

The user authorized the complete plan after the research checkpoint in section 45. P0–P6 are now locally complete for the named scope, as recorded in the final receipt. The new source-owned housing, commerce and ordinary-travel families have matching catalogue, direct, datetime and bounded-window REST surfaces; commerce additionally supports explicit buyer/seller instant comparison. VV and KP ordinary marriage extend the existing marriage surfaces with their own manifests, evidence, calendars, visibility and historical conditions.

The subsequent focused adversarial review's four findings are repaired: KP fixed-clock carryover, purpose-specific apparition admission, malformed shadow geometry and early VV history-budget admission. The repair acceptance records371 fresh passing cases, including a complete real KP HTTP window across the missed interval, strict REST failures and release/document gates.

Shared exact evidence computation requests only the selected dependencies. The three complete MC compatibility responses remain identical without normalization. Source fixtures, boundary/hostile-input tests, real DE441 HTTP requests, performance receipts and all release-facing document checks pass. Numerical certificates retain their declared-model scope and unresolved bands; work-budget failures never return complete partial results.

The selected feature count changes from eleven delivered/thirteen open to twelve delivered/twelve open. The broader VED-021/022 follow-through programmes remain open for future admissions. Door/site design, further financial acts, marine/military travel, disputed full KP travel, standalone BS/Raman/PVR marriage, regional rituals, compatibility and whole-ceremony windows remain explicit extensions outside this completed selection.

The prior source-published engine baseline is c0c1d739c9cdf52910ff2f64aea6ac0ad88e347f and wiki baseline is 7ad434a903555679a2393afd7487138bb9d3634c. These new changes are uncommitted and unpushed; version 6.9.9 is unchanged. No website/Urania delivery, release or deployment is claimed.

47. VED-011 four-area source publication — 11 October 2026

The user authorized commit and push of the complete implementation and all four focused-review repairs. The publication record identifies the scope, validation receipts and wiki-first publication order. The package includes all six generated wiki pages and the corresponding engine code, strict REST surfaces, tests, fixtures and canonical documents.

The original 501-case implementation acceptance and fresh 371-case repair acceptance retain their separate scopes. Pre-publication hashes confirm the tested code and fixtures are unchanged. The final commit identities belong to the engine/wiki histories; version 6.9.9, extension boundaries and the twelve delivered/twelve open feature count are unchanged. Release tags, website/Urania adoption and deployment are outside this publication.

48. VED-017 bounded Gochar windows — local completion, 11 October 2026

The user authorized all seven stages of the window implementation plan. The standard and final validation receipt record their completion: all seven subject/blocker streams, all twelve ayanamsas, direct and retrograde re-entries, complete conservative partitions, ordinary/baseline Vedha and optional once-derived raw natal BAV. Epoch/civil engine and facade calls, two typed REST calculation routes and resource-free capabilities share admission, numerical evidence, errors and exact JSON bytes.

The product admits up to 366 days inside its declared modern UT1/TT/resource intersection. Numerical width, contact, ordering and source uncertainty remain visible; strict requests reject unresolved results. It never substitutes a sampled grid for completeness or erases potential blocker events.

Acceptance accounts for 1,443 distinct tests across the documented runs, including all nine shared-owner real-history cases, plus 101 primary Horizons comparisons, 48 ERFA cases within the test set, source fixtures and 26 completed performance requests. The receipt preserves exact execution scopes, raw references, unchanged tolerances, repaired findings and the interrupted-run qualification. Annual latency and measured memory are documented rather than represented as an interactive performance promise.

The feature count changes to thirteen delivered / eleven open or bounded. VED-018/019 retain alternative sources, nodes, other references and modifiers; they are not missing requirements of this completed window profile. The wider VED-021/022 follow-through programmes remain open for future admissions. Changes are local and uncommitted/unpushed; version 6.9.9 is unchanged. Source publication, release and website/Urania adoption remain separate.

49. VED-017 source-publication package — 12 October 2026

The user authorized committing and pushing the complete bounded-window implementation, research, documentation and evidence. The publication checkpoint records reconciliation with upstream 7a86b33, retaining its additive API guards and all earlier changes. Exact byte checks preserve the implemented runtime and existing evidence. The fresh publication slice passes 803 tests with all 29 required DE441 cases executed and no failures or skips.

The generated wiki is published before the engine and its updated gitlink. Exact commit identities belong to the two repositories' histories. The thirteen-delivered/eleven-open count, source-selected boundaries and version 6.9.9 are unchanged. Source publication does not admit VED-018/019, release a new package, deploy the engine or change website/Urania behavior.

50. VED-012 research and implementation plan — 12 October 2026

The authorized online/local-library research and planning checkpoint is complete. The source packet separates JP 7.14's Navamsa-lord/movable-sign rule from Raman no. 252's D1-debilitation/D9-exaltation rule, with published Chart 122 as a sign-level witness. It collates all fourteen BPHS 41.2–15 wealth clauses and selects five chapter-39 source entries with explicit roles and relation witnesses.

The seven-stage plan covers source fixtures, shared strict admission, legacy non-evaluation metadata, exact D9 composition, the 21 rule entries, curated Python/facade and typed REST surfaces, adversarial validation and documentation. The existing-behavior baseline passes 195 tests with no failures or skips; read-only probes additionally establish direct/REST input discrepancies. That baseline does not validate the proposed new rules.

BPHS 39.35 remains explicitly non-admitted pending resolution of its aspect targets/type. An all-to-all full-graha-aspect reading is geometrically impossible; this is a specific interpretation problem, not source absence. Other chapter-39 domains remain separate extensions. How to Judge a Horoscope, Volume II, is available locally and online; the historical unnamed “Raman Part II” citation is not silently assigned to that book.

Runtime implementation has not begun. Changes at this checkpoint are the two canonical research/plan documents and this register; source publication is not part of this research action. The thirteen-delivered/eleven-open feature count and version 6.9.9 remain unchanged.

51. VED-012 selected source-rule closure - 12 October 2026

The user authorized execution of all seven stages. Four source profiles now admit 21 rule entries across module/root/facade/Vedic and typed REST surfaces. The standard fixes source identity, frames, quantifiers, directed aspects and non-evaluated context. The receipt records 573 passing tests without failures/skips, all 44 source fixtures through HTTP, independent D9/wealth/royal geometry, normal server startup, 144-chart legacy parity and reproducible cost measurements. Legacy computed fields are unchanged; the source-string update, strict admission and additive motion-state/policy transport have explicit migration notes.

The selected VED-012 package is locally complete, bringing the stable count to fourteen delivered / ten open-or-bounded. BPHS 39.35 remains research_required because its aspect targets/type need a source-owned resolution and satisfiable witness. Other chapter-39 domains and the old unidentified Raman citation remain source leads, not promised executable rules. No blanket classical chapter or predictive-accuracy certification is implied.

The eight touched engine files pass docstring preflight. A repository-wide run under the concurrently edited governance checker reported 23 issues in untouched house classes; the receipt separates that global condition from VED-012 acceptance. Existing concurrent governance, event-search/Gochar and light-box work is preserved. Generated Git-wiki synchronization, commit/push, release and deployment require their own publication action; none was performed here.

52. VED-012 source-publication package - 12 October 2026

The user authorized committing and pushing the completed VED-012 package. The publication checkpoint records the refreshed 573-test acceptance and passing full 366-file engine docstring preflight. The earlier house-class findings were repaired in the separate a399faa commit. Documentation, release identity, generated inventories and scoped lint checks pass. The generated wiki is published first; the parent engine commit containing this entry records its exact wiki gitlink. Concurrent light-box work is excluded. Version remains 6.9.9; source publication does not constitute a release or deployment.

The next numbered research package is VED-013: additional Chara Dasha lineages and source-owned Argala strength grading. Current Rao one/two-cycle Chara Dasha and count-based Argala remain admitted. Research must identify and compare named duration/direction/co-lord rules and any source-specific strength criteria before adding profiles. Shadbala integration is not an assumed universal grading rule. VED-013 implementation has not been started. The stable feature count remains fourteen delivered / ten open-or-bounded.

53. VED-013 online and local-library research - 12 October 2026

The source packet completes the authorized comparison of Chara variants and Argala grading. Local BPHS, Rao/Raman-attributed Jaimini material, Rao articles and Rangacharya's scanned manual were compared with primary author/publisher sources online. The manual's filename says 2009; its edition page identifies 2007. OCR absence was resolved by reading the scan, not treated as absence of the rules.

The manual supplies two Chara methods and a distinct female-chart recipe. Independent arithmetic reproduces 24 method-2 sequence rows (288 cells), seven of seven male example durations and eleven of twelve female example durations. The remaining Cancer entry conflicts with the page's own-sign rule. Other recorded decisions include zero-year treatment under dignity adjustment, co-lord precedence, source-specific subperiods and calendars. The PVR 2014 seed priority conflict and BPHS diagram/table limitations are retained explicitly; Narayana is a separately named family.

BPHS supplies one/two/three-plus Argala count grades, a three-malefic special threshold and directed degree-quarter obstruction qualifications. These are separate from planetary strength and from Rangacharya's Yogada grading. Node participation, reversal and equal-count comparison differ across sources. The packet identifies the required engine/REST contracts and confirms four invalid-input cases accepted directly but rejected through HTTP. Existing Argala's two-malefic/seven-occupant behavior remains unchanged pending repair.

The existing selected baseline passes 251 tests with no failures/errors/skips. Twelve direct/HTTP research probes and source arithmetic receipts are recorded separately; they do not validate an unimplemented profile. Only the source packet and this register change. VED-013 is RESEARCH_COMPLETE / PLAN_PENDING, with per-profile admission decisions still explicit. No runtime implementation, commit/push, release or deployment is included. The feature count remains fourteen delivered / ten open-or-bounded; version remains 6.9.9.

54. VED-013 selected implementation closure - 12 October 2026

The user authorized all seven stages after the research and implementation plan. They are now implemented and validated locally. This entry supersedes the implementation-not-started statements in the historical sections 52–53. The standard and validation receipt own the exact public contract and evidence.

The package supplies three immutable one-cycle Rangacharya arithmetic profiles, an optional separately attributed fixed-day projection, and three BPHS/PVR/Rath Argala assessments. Raw occupants, count grades, character, directed quarters, special-third roles and strength factors remain separate evidence. Discovery identifies executable, legacy and research-only records. Three new HTTP routes match public Python/facade behavior. Legacy valid arithmetic is unchanged; invalid direct/HTTP inputs now agree, and the legacy Argala receipt corrects its overbroad lineage claim.

Acceptance covers 847 distinct selected tests, with zero failures/errors/ skips in the reported passing runs. Printed sequences, partial source durations, the BPHS publisher erratum, PVR occupancy tables and synthetic exact boundaries remain distinct evidence classes. C04 retains female Cancer ten versus selected twelve. C12 records Rath's stated winner separately from Jupiter's symmetric rasi aspects to the two compared dual signs. Unavailable comparison doctrine produces an explicit indeterminate result, not a guessed winner. Validation, bounded-cost, real application lifecycle and generated-document receipts are recorded together; this is not whole-Vedic or predictive certification.

The remaining five admission gates are full BPHS Chara precedence/subperiods/ clock; PVR 2014 contradictions/strength thresholds/solar clock; separately owned Rath Narayana; Rangacharya Yogada with D3/D9 and merit aggregation; and a sourced general Argala strength comparator with group/tie/missing-context semantics. Their presence does not mean the selected seven-stage delivery is unfinished.

The current stable count is fifteen delivered / nine open-or-bounded. Canonical documentation and its generated mirror are prepared locally. No commit, push, version change, release, deployment or website work is included; concurrent light-box files are preserved.

55. VED-013 source-publication checkpoint - 12 October 2026

The user authorized commit/push of the selected six-profile implementation, then an adversarial review. The publication checkpoint defines the scoped manifest and refreshed acceptance gates. The generated wiki is published first; the containing parent engine commit records its exact gitlink. Concurrent light-box and natal-discovery work is excluded. The selected delivery count remains fifteen of 24, with nine open-or-bounded packages. Version remains 6.9.9; this is source publication, not a release or deployment. The subsequent adversarial review remains a separate evidence checkpoint.

56. VED-013 adversarial repairs - 12 October 2026

The authorized review of published 573570b identified three repairable defects, now fixed locally: Chara result witnesses were not fully bound to their selected recipe; Argala verdicts and derived evidence could contradict the chart; and constructor invariant failures could be reported as HTTP 422. The repair receipt records stronger whole-result validation and the post-admission internal-error boundary shared by engine construction and serialization. Client mistakes retain 422; corrupt engine evidence produces 500.

Acceptance passes 951 tests with zero failures/errors/skips, including 36,288 independently checked Chara duration rows, 2,640 Argala comparison cases, real server lifecycle and existing legacy/source/boundary tests. The six fixture files and all admitted source policies remain unchanged. Updated cost measurements and documentation guards pass with the stated concurrent server-edit boundary. The standard and generated wiki are reconciled locally.

The stable count remains fifteen delivered / nine open-or-bounded. The five VED-013 research candidates retain their admission gates. These repairs are not yet committed or pushed; the section-55 publication remains the last published VED-013 checkpoint. Concurrent natal-discovery/light-box work is preserved. Version remains 6.9.9; no release or deployment is included.

57. VED-013 repair publication and next numbered item - 12 October 2026

The user authorized commit/push of all three section-56 repairs. The publication receipt records unchanged source hashes against the 951-test acceptance, passing documentation/identity/inventory guards and the strictly scoped manifest. Generated wiki publication comes first; the containing engine commit records its exact gitlink. Concurrent natal-discovery/light-box work is excluded. This supersedes section 56's unpublished status without changing feature counts: fifteen delivered / nine open-or-bounded, version 6.9.9.

VED-014 - Kalachakra Dasha is the next numbered open feature. Its first step is source research online and in the local classical library for one identified edition/lineage: Savya/Apasavya traversal, Nakshatra/pada-to-sign tables, cycles, birth balance, year basis and nested periods. Source conflicts and independent worked fixtures must precede an engine/public/REST plan. Software comparisons can help detect differences but do not replace textual authority. This publication does not start or authorize that new package.

58. VED-014 Kalachakra source research - 12 October 2026

The user subsequently authorized deep research online and in C:\dev\ASTROLOGY-BOOKS-DATABASE. The research packet collates Santhanam's BPHS, Sastri's Jataka Parijata, Kapoor's Phaladeepika reflow, Rao's 2000 textbook and later formulation, Rath's 2005 handout, and software variant documentation. It identifies printed defects separately from substantive differences in balance, continuation, ADs, Deha/Jeeva and eligibility.

There are sufficient sources to plan a named first profile. The recommendation is Santhanam/BPHS longitude balance, natal-row repetition and cyclic natal-row ADs, with nine major-period occurrences and levels 1–2. The packet records all sixteen selected rows and 27 group memberships; independent rational arithmetic passed 468 assertions. Jataka Parijata's paired example distinguishes Scorpio with 3.09525 source years remaining from the pada-sign school's Libra with 10.428. Those calculations must not be collapsed into a universal default.

The next plan needs explicit source-correction decisions, year projection, occurrence identity, full/visible intervals, bounded active lookup and matched Python/facade/REST validation. Existing eight-lord alternate-dasha assumptions cannot be reused as Kalachakra doctrine. No new feature is implemented here; counts remain fifteen delivered / nine open-or-bounded. No publication, release, deployment or website work is included.

59. VED-014 selected Kalachakra implementation - 12 October 2026

The user authorized all seven stages of the implementation plan. The named bphs_santhanam_longitude_row_cycle_v1 profile is locally complete across engine, public exports, Moira methods and five REST endpoints. The validation receipt records 249 feature tests and 545 affected regressions, all passing without skips, including real DE441 date/direct/timezone and HTTP parity. Independent source fixtures were frozen before the calculator and retain their fingerprint.

Exact source-year intervals, occurrence paths, full-parent AD construction and explicit fixed-day projection prevent duplicate-sign selection and birth-clipping errors. Source-only results remain available if a short residual cannot be represented in projected JD. Discovery exposes all fourteen conflict decisions, unresolved Cancer/Leo classification and eligibility not evaluated. Alternative traditions remain distinct candidates; deeper levels, full gati classification and strength eligibility are separately scoped work.

This adds one selected delivery checkpoint: sixteen delivered / eight open or bounded. Python 3.10 compatibility is wired in CI but was not executed locally; the project Python 3.14.3 smoke passed. No commit, push, release, deployment or website work is included in this local completion.

60. Vedic release stopping point - 12 October 2026

The user selected Moira 6.10.0 and authorized the final review, repairs, integrated validation, source publication and engine release. Feature expansion pauses at the selected VED-014 checkpoint: sixteen delivered / eight open or bounded. No additional numbered package is started to complete this release.

The release receipt records the Kalachakra error-classification repair, historical catalog-gate repair and release validation. The release notes and compatibility notes define the shipped scope. Separate unfinished natal-discovery/light-box work is excluded; website application deployment remains separate. Earlier local-only statements in sections 58-59 describe those historical checkpoints.