# Stage 149: 4-Stage Scenario Funnel & J(s) (aq149_filter_trillion_scenarios_and_jagged_frontier / DB-UNIT-18)

Filters massive combinatorial scenario streams through a 4-stage funnel and ranks survivors by Jagged-Frontier variance and model disagreement J(s).

- Atomic micro-queue: `aq149_filter_trillion_scenarios_and_jagged_frontier` (visibility: `public_commons`)
- Backing database unit: `DB-UNIT-18` (`Causal Scenario Foundry, Simulation Federation & Opportunity Core`)
- Database cell: `entities-db.teleon.dev` (`teleon_core.scenario_ir_manifests`, `teleon_core.world_hidden_truth_vault`, `teleon_core.actor_observation_views`, `teleon_core.opportunity_hypotheses`)
- Claim row hydration: `single_self_contained_flat_task_row` (100% context inline, normalized writeback)
- Track 2 Sister queue (`SISTER-18`): `sq18-improve-scenarios-queue.teleon.dev` (Enforce POET minimal-criterion solvability bands, audit SPDX simulator licenses, and turn production failures into regression worlds.)
- Lanes: Hard Invariant Prune, Novelty Dedup, Jagged Frontier J(s)

## Execution engines & AI Agent freedom contract

- **Deterministic (0 model calls)**: `mesa_amlsim_and_sql_state_world_compiler_v1` (4-stage constraint/MinHash/budget funnel & J(s) scorer (0 calls))
- **Hybrid (<=1 model call)**: `autoplay_explore_first_scenario_synthesizer_v1`
- **AI Agent & Bot Fleet (1..6 calls)**: `multi_engine_concordia_oasis_and_foresight_author_v1` (High-J(s) frontier cliff & failure-mode probe author)
  - Contract: **Instructions + Resources + Freedom to search for new resources and new ways not bound by instructions**

## Database invariants

- Investigator and solver views never access world_hidden_truth_vault or derive test fixtures from detector rules.
- Every suspicious or failure scenario pairs with a matched benign counterfactual and abstention control.
- Opportunity and problem generation stays decoupled from pre-selected solution promotion.

## External & internal bot fleet actions

- Prune physically impossible and near-duplicate scenarios at Stage 1-2 in 0 model calls
- Score Stage 3-4 candidates by neighbor cliff sharpness and cross-model disagreement J(s)
- Route top frontier anomalies directly into targeted distillation and shim queues

## Sister self-tuning areas

- Enforce POET solvability bands and family-level train/holdout lineage splits (DIM-01, DIM-02)
- Wrap permissive simulation adapters (MIT/Apache-2.0/BSD-3) while blocking copyleft or BSL traps (DIM-03)
- Convert verified production traces and failed experiments into permanent regression scenarios (DIM-06, DIM-07)