docs(skill): publish latest truth dashboards

2026-06-27 00:10:33 +08:00
parent a16a157f0a
commit adcdb7e987
3 changed files with 179 additions and 0 deletions
@@ -0,0 +1,58 @@
+# Jyotish External Oracle Closure Master Dashboard
+
+Generated: `2026-06-26T16:09:33.650704+00:00`
+
+## Summary
+
+- total_tasks: `12`
+- external_verified_tasks: `0`
+- open_tasks: `12`
+- can_claim_global_oracle_closure: `false`
+- production_tuning_allowed: `false`
+
+## Fronts
+
+| front | tasks | verified | first priority | missing fields | manual entries | metadata missing | target missing |
+|---|---:|---:|---|---:|---:|---:|---:|
+| `dasha` | 3 | 0 | `template_steve_jobs_dasha_lahiri` | 6 | 6 | 5 | 1 |
+| `tajika_sahams` | 5 | 0 | `template_steve_jobs_varshaphala_1984_lahiri` | 15 | 15 | 5 | 10 |
+| `shadbala` | 4 | 0 | `template_redacted_place_shadbala_raman` | 55 | 55 | 5 | 50 |
+
+## Next Action Order
+
+### dasha
+
+- case_id: `template_steve_jobs_dasha_lahiri`
+- capture_id: `external_template_steve_jobs_dasha_lahiri`
+- missing_field_count: `6`
+- manual_entry_count: `6`
+
+```bash
+python3 scripts/oracle_collection_queue.py --oracle-file references/oracle/dasha_shadbala_oracle_cases.json --apply-packet references/oracle/artifacts/pending_packets/external_template_steve_jobs_dasha_lahiri.json --format json
+```
+
+### tajika_sahams
+
+- case_id: `template_steve_jobs_varshaphala_1984_lahiri`
+- capture_id: `external_template_steve_jobs_varshaphala_1984_lahiri`
+- missing_field_count: `15`
+- manual_entry_count: `15`
+
+```bash
+python3 scripts/tajika_annual_oracle_queue.py --oracle-file references/oracle/tajika_annual_oracle_cases.json --apply-packet references/oracle/artifacts/pending_packets/external_template_steve_jobs_varshaphala_1984_lahiri.json --format json
+```
+
+### shadbala
+
+- case_id: `template_redacted_place_shadbala_raman`
+- capture_id: `external_template_redacted_place_shadbala_raman`
+- missing_field_count: `55`
+- manual_entry_count: `55`
+
+```bash
+python3 scripts/oracle_collection_queue.py --oracle-file references/oracle/dasha_shadbala_oracle_cases.json --apply-packet references/oracle/artifacts/pending_packets/external_template_redacted_place_shadbala_raman.json --format json
+```
+
+## Boundary
+
+This dashboard merges external evidence readiness only. It does not claim prediction accuracy, does not tune production constants, and does not treat local engine output as oracle evidence.
@@ -0,0 +1,38 @@
+# Public Jyotish Benchmark Dashboard
+
+Generated: `2026-06-26T16:09:33.617633+00:00`
+
+## Capability Registry
+
+- technique_count: `79`
+- capability_valid: `true`
+- problem_count: `0`
+
+## Dasha/Shadbala Oracle Readiness
+
+- total_packets: `5`
+- valid_packets: `0`
+- ready_for_calibration: `0`
+- production_tuning_allowed: `false`
+
+## Boundary Audit
+
+- external_verified_template_cases: `0`
+- template_comparison_count: `0`
+- production_tuning_recommended: `false`
+
+## Global First Claim
+
+- can_claim_global_first: `false`
+- reason: Do not claim global first until Dasha/Shadbala external oracle packets are valid, production tuning is allowed, and public benchmark history is stable.
+
+## Remaining Gap
+
+Dasha/Shadbala external oracle readiness remains 0, Shadbala absolute values still need component-level external evidence, and public long-term benchmark history is not yet comparable to the strongest global open-source projects.
+
+## Next Actions
+
+- Fill the first external JHora/PyJHora packet under references/oracle/artifacts/pending_packets.
+- Run oracle_evidence_validator.py until at least one packet is valid.
+- Run oracle_boundary_audit.py to inspect Dasha/Shadbala deltas without tuning constants.
+- Publish this dashboard after each validated sample batch.
@@ -0,0 +1,83 @@
+# Jyotish Skill Gap Truth Audit
+
+Generated: `2026-06-26T16:09:34.006081+00:00`
+
+## Public Claim Boundary
+
+- can_claim_global_first: `false`
+- can_claim_all_skills_complete: `false`
+- can_claim_perfect_accuracy: `false`
+- reason: The skill has strong breadth, but external oracle closure, long-term benchmark history and several traditional judgment layers remain unfinished.
+
+## Capability Snapshot
+
+- technique_count: `79`
+- capability_valid: `true`
+- hard_front_count: `5`
+- past_correction_count: `6`
+
+## External Oracle Closure
+
+- total_tasks: `12`
+- external_verified_tasks: `0`
+- open_tasks: `12`
+- can_claim_global_oracle_closure: `false`
+
+## Remaining Hard Fronts
+
+### Dasha external oracle
+
+- id: `dasha_external_oracle`
+- priority: `P0`
+- status: `blocked_external_evidence`
+- current_truth: Dasha engines are usable, but exact start dates, balance periods and sub-period boundaries cannot be called externally closed until real JHora/PyJHora/book-example packets pass validation.
+
+### Long-term public benchmark
+
+- id: `long_term_public_benchmark`
+- priority: `P0`
+- status: `active_gap`
+- current_truth: Benchmark scripts exist, but the skill still lacks a long-running public history comparable to the strongest global projects.
+
+### Shadbala external absolute values
+
+- id: `shadbala_external_absolute_values`
+- priority: `P0`
+- status: `blocked_external_evidence`
+- current_truth: Internal six-component Rupa/Virupa aggregation is usable, but external absolute-value closure is not complete.
+
+### Tajika / Sahams annual closure
+
+- id: `tajika_sahams_annual_closure`
+- priority: `P0`
+- status: `active_gap`
+- current_truth: Varshaphala, Muntha, Year Lord, Mudda Dasha and Sahams have local structure, but annual judgment depth and external sample closure remain incomplete.
+
+### Article-level interpretation template industrialization
+
+- id: `article_template_industrialization`
+- priority: `P1`
+- status: `active_gap`
+- current_truth: Several high-granularity templates are frozen, but article-derived rules must keep being distilled into guarded templates with sources, cross-checks and forbidden claims.
+
+## Past Corrections
+
+- `ashtakoot_not_all_zero`: Ashtakoot matrices and non-zero scoring exist; the remaining issue is external validation and finer-grained comparison, not total absence.
+- `panchanga_not_empty`: Panchanga and Muhurta engines are deep; remaining work is UI, external oracle comparison and festival/v vrata breadth.
+- `covered_is_not_complete`: Covered means there is a stable entry or usable layer; complete requires mature workflow, tests, boundary text and, where needed, external oracle closure.
+- `single_factor_reading_risk`: Use D1, relevant Varga, Dasha, Transit, strength and historical verification where applicable. Single-factor readings should be capped at low confidence.
+- `enemy_sign_not_debilitation`: Enemy sign and debilitation are separate dignity states. Do not label a planet as debilitated unless it is in the classical debilitation sign/degree zone.
+- `jaimini_kp_tajika_not_absent`: These fronts have usable structures; the hard work is tail-difference closure, traditional workflow deepening and external sample validation.
+
+## Must Not Overclaim
+
+- Dasha exact-date boundaries are externally closed.
+- Shadbala absolute Rupa values are fully calibrated against JHora or book examples.
+- Tajika/Sahams annual prediction is traditional-software-grade closed.
+- All Indian astrology techniques are complete.
+- The skill is already the undisputed global open-source number one.
+- A single chart factor, article rule, screenshot or template can override D1/Varga/Dasha/Transit convergence.
+
+## Boundary
+
+This audit answers the skill-level truth question. A covered technique is not the same as a complete externally closed technique; single-factor case readings must remain confidence-capped.