Chart-driven detection of 48 classical Vedic yogas. Twelve conjunction and dignity yogas: Gajakesari (parashara three-rule definition), Sunapha, Anapha, Dhurdhura, Kemadruma, Chandra Mangala, Budha-Aditya, and the five Pancha Mahapurusha yogas (Ruchaka, Bhadra, Hamsa, Malavya, Sasa). Plus all 32 Nabhasa distribution yogas, which describe how the seven visible grahas are spread across the whole chart rather than any single conjunction, across four families: Asraya (Rajju, Musala, Nala), Dala (Mala, Sarpa), Akriti (Gada, Shakata, Vihaga, Shringataka, Hala, Vajra, Yava, Kamala, Vapi, Yupa, Shara, Shakti, Danda, Nauka, Kuta, Chhatra, Dhanusha, Ardhachandra, Chakra, Samudra) and Sankhya (Gola, Yuga, Shoola, Kedara, Pasa, Damini, Veena). Plus four wealth and poverty verdicts, each ONE answer over a whole family of classical rules: Dhana Yoga over the eleven catalogued wealth combinations of BPHS ch. 41, Daridra Yoga over the poverty combinations of BPHS ch. 42 and Phaladeepika ch. 6, Lakshmi Yoga (BPHS ch. 36), and Dhana Malika (Jataka Parijata ch. 7). Their evidence names every rule that matched and the exact condition it matched on, so a wealth reading cites the combination rather than a label, and a rule resting on a single authority is excluded from the verdict and says so rather than quietly counting. Each yoga is returned with an `id`, `name`, a `present` boolean, a `quality` (Positive, Negative, or Both, i.e. auspicious, inauspicious, or context-dependent), and a classical-text `evidence` string naming the rule that triggered or failed (kendra position, dignity, malefic drishti, lordship, retrograde state, sign modality, bhava distribution). Nabhasa results also apply the four classical precedence norms, so a yoga that matched its own rule but was outranked by a stronger family is returned as absent with evidence naming the norm that silenced it, letting you explain a verdict rather than only report it. There is no separate major/minor flag; `quality` is the auspiciousness axis. Unlike GET /yoga and GET /yoga/{id} which are dictionary lookups, this endpoint computes the kundli from birth data and runs the detection rules. Sources: BPHS ch. 35 and ch. 75, Mantreswara Phaladeepika ch. 6, B.V. Raman Three Hundred Important Combinations.