commonground-scenarios

Deterministic, offline scenario generation for the Common Ground environment family. Each generated scenario contains a fictional organization, policy documents, planted ambiguities/contradictions/gaps, precise distractors, a seeded persona panel, and provenance.

Every canonical clarifying question is phrased as a yes/no proposition. Its target_stances have an explicit interpretation: agree predicts yes, disagree predicts no, and pass means the faction takes no position. Templates may also author a finite canonical_question_aliases allowlist. Aliases obey the same yes/no form and are kept distinct across the answer key.

The default generator uses committed templates only. An operator may inject a prose-polishing callback, but no model client or network path is built into the package, and validation requires every planted anchor to survive polishing.

Training and held-out template registries and template/layout-profile labels are disjoint. The canonical release contains 100 training and 100 held-out rows. The generator assigns opaque document/faction IDs, varies visible layout and faction count/order, and records decision terms, answer-conditioned value, and paired evidence for contradictions. Split generation distinguishes exact instance, canonical prompt, and policy-issue fingerprints; it also runs token-Jaccard and word-ngram TF-IDF cross-split neighbor audits. Generated scenarios are canonical JSON and regenerate byte-for-byte from the same template, seed, and explicit generation date.

The bundled scenarios are synthetic. The optional human_feedback field supports advanced custom input and delegates to the shared fail-closed validator. It is not used by the bundled data; see human-data governance for its separate review requirements.

The packaged Draft 2020-12 schema has the stable identifier urn:commonground:schema:scenario:2 and is available through load_scenario_schema.