Replace the Direct Agentic textStream relay with a durable V9 agent runtime:
- agentic-rectification.ts: short boundary-only system prompt (no gate->scan
->score->diagnostics copy); pins skills/jyotish-birth-time-rectification;
per-action bounded maxSteps (opening/read-only 6, evidence 8, rescore 12,
accept/confirm 6) with a hard ceiling and repeated-tool-call detection.
- rectification-v9-tools.ts: ten Case-ref tools (read-case, propose/confirm/
revise-evidence, compare-candidates, read-diagnostics, offer-candidates,
accept-candidate, confirm-birth-time, close-case). Inputs are minimal refs
only; RPC-backed evidence ledger, fingerprint cache reuse, receipts, and
accepted!=confirmed semantics; confirm requires gate + grounded consent.
- /api/rectification/agent: caseId/sessionId/requestId/action/message; exact
Case<->Session binding verified server-side; client history never overrides
the durable dossier; pending turn -> completed/failed/retryable; consumes
result.fullStream and emits allowlisted NDJSON only (reasoning/raw/provider
metadata/tool payloads/birth data/scores never forwarded); first-turn real
skill.started/skill.loaded gate with one controlled retry; billing bound to
rectification:case:{caseId}.
- New forward migration 20260813010000_agentic_rectification_v9_agent_api.sql:
case dossier/compute, turn finalize, fingerprint-cached candidate persist,
case-scoped accept, consent-gated confirm, guarded transitions,
needs_rebaseline profile guard, run_phases receipt table, and the
rectification_runtime_version feature flag (v9 default, legacy read-only).
- Frontend: homepage/sidebar entry routing now uses the server Case open API
(openRectificationFromHomepage/openRectificationSession/startNewRectification)
with exact sessionId/caseId and server-owned shouldStartOpening; CTA driven
by entry-summary; chat restores from persisted turns, candidate cards from
the Candidate Snapshot API, activity from real NDJSON + persisted receipts;
direct durable candidate-accept endpoint for the UI cards.
The recovery migration crossed the identity and RBAC ledgers without guarding schema prerequisites, while unknown configuration, provider, and database failures escaped the admin authorization boundary as 500s. Keep recovery in the DB ledger with explicit prerequisite no-ops, and sanitize unknown authorization failures to the existing 503 path.
Owner recovery lived in the identity-only migration path and could clear historical revocations on conflict. Admin layout also rethrew authorization 503s, turning service-unavailable failures into 500 responses.\n\nMove recovery behind the RBAC migration sequence, exclude revoked candidates without mutating their history, and terminate layout redirects at a no-store 503 route.
Add a new agentic rectification flow that lets an LLM drive the full
jyotish-vedic-astrology methodology on the web, with the Python engine as
its computation layer (mirroring local Claude Code):
- mastra/rectification-tools.ts: 7 engine tools (gate/scan/score/diagnostics/
candidate-features/confirm/save-birth-time)
- mastra/agentic-rectification.ts: agent mounting the full skill + tools
- lib/rectification-agentic/session.ts: server-owned profile + confirmation
gate; the LLM can only persist the exact minute the engine's high-rigor
gate confirmed
- app/api/rectification/agent/route.ts: NDJSON streaming endpoint with
credit reserve/settle
- components/rectification-agentic-chat.tsx + entry switch: new sessions use
the agentic chat; in-progress v4 cases still resume on the v4 panel
- migration 20260801000000: service-role RPC writing profiles.active_birth_time
with baseline concurrency guard
- tests for tools + session (12 cases); full suite passes 1076
Co-Authored-By: Claude <noreply@anthropic.com>
Jesse_Chen