prior-accounting_firmsrecord 1
{
"id": "prior-accounting_firms",
"industry": "accounting_firms",
"observed": "2026-08-27",
"evidence_class": "E/I",
"source": "research/actionmodel-builder-research-2026-08-26/phase-2/outputs/industry-atom-specifications.md#4.1",
"claim": "Accounting firms' painful jobs center on document chasing, bank-feed reconciliation and period close; core entities and authority boundaries are known before any client conversation.",
"atoms": [
"intake_normalize",
"extract_structure",
"reconcile_audit",
"follow_up_chase",
"report_digest"
],
"entities": [
"client",
"engagement",
"period",
"document",
"bank_transaction",
"ledger_entry",
"invoice",
"exception"
],
"terminal_owner": "controller_or_engagement_bookkeeper",
"precomputable": [
"entity vocabulary and period/close cycle",
"likely systems of record (GL, bank feeds, document store)",
"archetype fit (case/workflow + reconciliation desk)",
"authority boundary defaults (no posting, no write-off)"
],
"must_ask": [
"which GL/bank-feed/document systems and who owns the period",
"close calendar and materiality thresholds",
"which exceptions actually cost hours (document chase vs matching vs review)",
"client volume denominators (clients, transactions/month)"
],
"contradiction_risks": [
"stated want 'automate the close' vs observed pain usually document collection",
"source-of-truth ambiguity between GL and bank feed"
],
"discovery_confidence_note": "demand_signal=E, validated_demand=U; priors are catalogue+spec evidence, not interviews",
"limitations": "No validated industry denominator; system-of-record mix varies by firm size and country",
"disposition": "prior_retained"
}
prior-constructionrecord 2
{
"id": "prior-construction",
"industry": "construction",
"observed": "2026-08-27",
"evidence_class": "E/I",
"source": "research/actionmodel-builder-research-2026-08-26/phase-2/outputs/industry-atom-specifications.md#4.2",
"claim": "Construction demand centers on job tracking, change-order chase, subcontractor coordination and progress billing; multi-party authority is the defining constraint.",
"atoms": [
"intake_normalize",
"triage_route",
"sync_handoff",
"follow_up_chase",
"report_digest"
],
"entities": [
"project",
"job",
"milestone",
"change_order",
"subcontractor",
"progress_claim",
"invoice",
"approval",
"exception"
],
"terminal_owner": "project_manager_or_commercial_manager",
"precomputable": [
"multi-party actor model (GC, subs, client, certifier)",
"change-order and progress-claim state machines",
"document-versioning requirement",
"archetype fit (case/workflow + field ops)"
],
"must_ask": [
"which PM/schedule system is authoritative and which spreadsheets shadow it",
"approval chain for change orders (who, thresholds)",
"subcontractor communication channels and evidence formats",
"billing cadence and retention rules"
],
"contradiction_risks": [
"schedule truth split across PM tool and site spreadsheets",
"claimed approval process vs actual email/phone approvals"
],
"discovery_confidence_note": "demand_signal=E, validated_demand=U",
"limitations": "Corpus has weak construction-specific supply; priors are demand-side only",
"disposition": "prior_retained"
}
prior-course_creatorsrecord 3
{
"id": "prior-course_creators",
"industry": "course_creators",
"observed": "2026-08-27",
"evidence_class": "E/I",
"source": "research/actionmodel-builder-research-2026-08-26/phase-2/outputs/industry-atom-specifications.md#4.3",
"claim": "Course creators need student onboarding, lesson drips and community management; learner identity and consent are the sensitive boundary.",
"atoms": [
"intake_normalize",
"schedule_coordinate",
"follow_up_chase",
"report_digest"
],
"entities": [
"learner",
"cohort",
"enrollment",
"lesson",
"attendance_event",
"question",
"content_item",
"consent",
"exception"
],
"terminal_owner": "course_operator_or_cohort_lead",
"precomputable": [
"cohort/enrollment/lesson state machines",
"common platform set (LMS, community, payment)",
"archetype fit (learning/content + portal)"
],
"must_ask": [
"which LMS/community/payment platforms and their data-access reality",
"cohort cadence and drip logic",
"what 'stalled learner' means to this operator",
"consent posture for learner contact"
],
"contradiction_risks": [
"stated want 'more automation' vs pain usually unanswered questions and stalled cohorts",
"platform data often locked in closed SaaS"
],
"discovery_confidence_note": "demand_signal=E, validated_demand=U",
"limitations": "Solo-operator vs team scale changes everything; no denominator",
"disposition": "prior_retained"
}
prior-ecommercerecord 4
{
"id": "prior-ecommerce",
"industry": "ecommerce",
"observed": "2026-08-27",
"evidence_class": "E/I",
"source": "research/actionmodel-builder-research-2026-08-26/phase-2/outputs/industry-atom-specifications.md#4.4",
"claim": "Ecommerce demand centers on cart recovery, inventory sync and support tickets; payment/PII and order authority are the hard boundaries.",
"atoms": [
"monitor_alert",
"follow_up_chase",
"sync_handoff",
"triage_route",
"report_digest"
],
"entities": [
"customer",
"cart",
"order",
"line_item",
"product",
"inventory_location",
"ticket",
"payment_status",
"exception"
],
"terminal_owner": "ecommerce_operations_or_support_lead",
"precomputable": [
"platform-detectable stack (storefront, payment, support, 3PL) from public site",
"order/inventory/ticket state machines",
"archetype fit (support + inventory ops)",
"stack detection via public storefront fingerprinting is unusually strong here"
],
"must_ask": [
"inventory source of truth across locations/channels",
"SLA targets and ticket taxonomy",
"recovery-message consent and brand voice",
"peak-season constraints"
],
"contradiction_risks": [
"'inventory sync' stated as want while root cause is multi-channel ownership ambiguity",
"support metrics vs actual queue reality"
],
"discovery_confidence_note": "demand_signal=E, validated_demand=U",
"limitations": "Storefront fingerprinting shows stack, not workflow pain",
"disposition": "prior_retained"
}
prior-education_trainingrecord 5
{
"id": "prior-education_training",
"industry": "education_training",
"observed": "2026-08-27",
"evidence_class": "E/I",
"source": "research/actionmodel-builder-research-2026-08-26/phase-2/outputs/industry-atom-specifications.md#4.5",
"claim": "Education/training demand centers on enrolment, attendance, learner communication and tuition administration; guardian/consent authority is industry-specific.",
"atoms": [
"intake_normalize",
"schedule_coordinate",
"follow_up_chase",
"reconcile_audit"
],
"entities": [
"learner",
"guardian_or_sponsor",
"program",
"enrollment",
"session",
"attendance",
"invoice",
"payment",
"consent",
"exception"
],
"terminal_owner": "program_administrator_or_training_coordinator",
"precomputable": [
"enrolment/attendance/tuition state machines",
"guardian/sponsor actor model",
"archetype fit (scheduling + portal + reconciliation)"
],
"must_ask": [
"SIS/LMS and tuition-ledger systems and identity authority",
"session model (cohort, rolling, 1:1) and timezone spread",
"guardian communication rules",
"funding/sponsor reporting obligations"
],
"contradiction_risks": [
"attendance recorded in multiple places",
"tuition aging blamed on parents vs actually on invoicing gaps"
],
"discovery_confidence_note": "demand_signal=E, validated_demand=U",
"limitations": "Regulatory variation by jurisdiction unknown",
"disposition": "prior_retained"
}
prior-healthcare_medical_practicesrecord 6
{
"id": "prior-healthcare_medical_practices",
"industry": "healthcare_medical_practices",
"observed": "2026-08-27",
"evidence_class": "E/I",
"source": "research/actionmodel-builder-research-2026-08-26/phase-2/outputs/industry-atom-specifications.md#4.6",
"claim": "Healthcare practice demand is administrative (appointments, no-shows, reminders, billing admin); clinical data and decisions are excluded by boundary, which shapes discovery itself.",
"atoms": [
"schedule_coordinate",
"follow_up_chase",
"triage_route",
"reconcile_audit"
],
"entities": [
"patient_or_subject",
"appointment",
"practitioner",
"clinic",
"attendance",
"reminder_consent",
"invoice",
"payer",
"exception"
],
"terminal_owner": "practice_administrator",
"precomputable": [
"admin-vs-clinical boundary (clinical record is a prohibited source)",
"appointment/billing state machines",
"archetype fit (scheduling + admin desk)",
"discovery must itself avoid ingesting clinical data"
],
"must_ask": [
"PMS/billing systems and payer mix",
"no-show economics and reminder consent posture",
"which admin tasks consume staff hours",
"data-boundary conformance requirements (jurisdictional)"
],
"contradiction_risks": [
"stated 'reduce no-shows' vs actual pain often billing exceptions",
"consent state scattered"
],
"discovery_confidence_note": "demand_signal=E, validated_demand=U; synthetic-only pilots per prior spec",
"limitations": "Regulated-data constraints make evidence-based discovery (scraping, log mining) mostly unavailable; question-led discovery dominates here",
"disposition": "prior_retained"
}
prior-hospitalityrecord 7
{
"id": "prior-hospitality",
"industry": "hospitality",
"observed": "2026-08-27",
"evidence_class": "E/I",
"source": "research/actionmodel-builder-research-2026-08-26/phase-2/outputs/industry-atom-specifications.md#4.7",
"claim": "Hospitality demand centers on reservations, guest feedback and housekeeping coordination; timing conflicts are the recurring exception class.",
"atoms": [
"schedule_coordinate",
"triage_route",
"sync_handoff",
"monitor_alert"
],
"entities": [
"guest",
"reservation",
"room",
"stay",
"housekeeping_task",
"maintenance_issue",
"feedback",
"exception"
],
"terminal_owner": "front_office_or_housekeeping_manager",
"precomputable": [
"stay/task state machines",
"PMS-centric system model",
"archetype fit (scheduling + ops coordination)",
"property size/type inferable from public listings"
],
"must_ask": [
"PMS and channel-manager reality",
"housekeeping task ownership and shift model",
"feedback channels and escalation rules",
"seasonality and occupancy patterns"
],
"contradiction_risks": [
"room-status disagreement between PMS and housekeeping reality",
"feedback treated as authority when it is evidence"
],
"discovery_confidence_note": "demand_signal=E, validated_demand=U",
"limitations": "Independent property vs chain changes system landscape entirely",
"disposition": "prior_retained"
}
prior-it_services_mspsrecord 8
{
"id": "prior-it_services_msps",
"industry": "it_services_msps",
"observed": "2026-08-27",
"evidence_class": "E/I",
"source": "research/actionmodel-builder-research-2026-08-26/phase-2/outputs/industry-atom-specifications.md#4.8",
"claim": "MSP demand centers on ticket triage, SLA/backup monitoring and client onboarding; privileged access and secrets are the hard boundary.",
"atoms": [
"triage_route",
"monitor_alert",
"intake_normalize",
"report_digest",
"approval_publish"
],
"entities": [
"client",
"tenant",
"ticket",
"incident",
"SLA",
"backup_job",
"asset",
"onboarding_item",
"secret_reference",
"exception"
],
"terminal_owner": "service_desk_lead_or_client_success_owner",
"precomputable": [
"ITSM-centric system model (PSA/RMM vocabulary)",
"SLA/ticket state machines",
"multi-tenant client model",
"archetype fit (support desk + monitoring)",
"MSPs are unusually legible: PSA/RMM tools are a small known set"
],
"must_ask": [
"which PSA/RMM and their tenant hygiene",
"SLA definitions per client tier",
"onboarding checklist reality",
"escalation and after-hours model"
],
"contradiction_risks": [
"SLA clocks computed differently across tools",
"tenant mix-ups in shared tooling"
],
"discovery_confidence_note": "demand_signal=E, validated_demand=U",
"limitations": "Secrets/privileged access excluded from any automated discovery",
"disposition": "prior_retained"
}
prior-insurance_agenciesrecord 9
{
"id": "prior-insurance_agencies",
"industry": "insurance_agencies",
"observed": "2026-08-27",
"evidence_class": "E/I",
"source": "research/actionmodel-builder-research-2026-08-26/phase-2/outputs/industry-atom-specifications.md#4.9",
"claim": "Insurance agency demand centers on lead follow-up, renewal chasing and certificate issuance; carrier record is authoritative and licensure bounds authority.",
"atoms": [
"follow_up_chase",
"extract_structure",
"approval_publish",
"report_digest"
],
"entities": [
"prospect",
"policy",
"insured",
"renewal",
"condition",
"certificate_request",
"broker",
"document",
"consent",
"exception"
],
"terminal_owner": "licensed_broker_or_agency_ops_lead",
"precomputable": [
"policy/renewal state machines",
"AMS-centric system model",
"carrier-vs-agency authority split",
"archetype fit (CRM + document workflow)"
],
"must_ask": [
"AMS/CRM reality and carrier portal mix",
"renewal book size and cycle timing",
"certificate volume and turnaround expectations",
"licensure boundaries on who may communicate what"
],
"contradiction_risks": [
"renewal date truth split between AMS and carrier",
"certificate 'urgent' claims vs actual issuance authority"
],
"discovery_confidence_note": "demand_signal=E, validated_demand=U",
"limitations": "Regulated advice boundary; jurisdictional variation unknown",
"disposition": "prior_retained"
}
prior-law_firmsrecord 10
{
"id": "prior-law_firms",
"industry": "law_firms",
"observed": "2026-08-27",
"evidence_class": "E/I",
"source": "research/actionmodel-builder-research-2026-08-26/phase-2/outputs/industry-atom-specifications.md#4.10",
"claim": "Law firm demand centers on client intake, conflict checks, document drafting and billing admin; privilege classification constrains every data flow including discovery itself.",
"atoms": [
"intake_normalize",
"triage_route",
"extract_structure",
"approval_publish",
"follow_up_chase"
],
"entities": [
"prospective_client",
"matter",
"party",
"conflict_hit",
"document",
"attorney",
"invoice",
"privilege_class",
"exception"
],
"terminal_owner": "responsible_lawyer_or_conflicts_manager",
"precomputable": [
"matter/intake/conflict state machines",
"practice-management system vocabulary",
"privilege boundary (discovery may only touch metadata)",
"archetype fit (case/workflow + intake desk)"
],
"must_ask": [
"PMS/DMS/billing systems and matter numbering authority",
"practice-area mix (drives entity variants)",
"conflict-check process reality",
"intake volume and conversion economics"
],
"contradiction_risks": [
"intake speed wanted vs conflict-check rigor required",
"billing aging blamed on clients vs internal WIP release"
],
"discovery_confidence_note": "demand_signal=E, validated_demand=U",
"limitations": "Privilege prevents content-level evidence mining; metadata-only discovery",
"disposition": "prior_retained"
}
prior-logistics_freightrecord 11
{
"id": "prior-logistics_freight",
"industry": "logistics_freight",
"observed": "2026-08-27",
"evidence_class": "E/I",
"source": "research/actionmodel-builder-research-2026-08-26/phase-2/outputs/industry-atom-specifications.md#4.11",
"claim": "Logistics demand centers on shipment exceptions, carrier booking, POD filing and customer notices; event ordering and idempotency dominate the data model.",
"atoms": [
"monitor_alert",
"triage_route",
"schedule_coordinate",
"sync_handoff",
"follow_up_chase"
],
"entities": [
"shipment",
"load",
"carrier",
"stop",
"tracking_event",
"delivery_window",
"proof_of_delivery",
"customer",
"exception"
],
"terminal_owner": "transport_operations_or_logistics_coordinator",
"precomputable": [
"shipment/event state machines with out-of-order handling",
"TMS/carrier-tracking system model",
"archetype fit (exception desk + coordination)"
],
"must_ask": [
"TMS and tracking-source mix; where POD actually lives",
"exception taxonomy that costs money (late, damaged, missing POD)",
"customer notification obligations",
"carrier mix and EDI/API reality"
],
"contradiction_risks": [
"ETA truth conflicts across sources",
"'real-time visibility' wanted vs data arrives batched/late"
],
"discovery_confidence_note": "demand_signal=E, validated_demand=U",
"limitations": "Corpus supply thin for logistics; demand priors only",
"disposition": "prior_retained"
}
prior-marketing_social_media_agenciesrecord 12
{
"id": "prior-marketing_social_media_agencies",
"industry": "marketing_social_media_agencies",
"observed": "2026-08-27",
"evidence_class": "E/I",
"source": "research/actionmodel-builder-research-2026-08-26/phase-2/outputs/industry-atom-specifications.md#4.12",
"claim": "Agency demand centers on content scheduling, reporting decks and retainer-lead chasing; client approval is the authoritative gate for publishing.",
"atoms": [
"approval_publish",
"schedule_coordinate",
"report_digest",
"follow_up_chase",
"extract_structure"
],
"entities": [
"client",
"campaign",
"content_item",
"channel",
"approval",
"metric_snapshot",
"lead",
"retainer",
"asset_right",
"exception"
],
"terminal_owner": "client_account_lead_with_named_client_approver",
"precomputable": [
"campaign/approval state machines",
"channel/analytics system vocabulary",
"multi-client tenancy model",
"archetype fit (approval workflow + reporting)",
"agency's own public portfolio reveals client mix"
],
"must_ask": [
"which scheduling/analytics/CRM tools per channel",
"approval chain per client and turnaround expectations",
"reporting cadence and what clients actually read",
"asset-rights handling"
],
"contradiction_risks": [
"'save reporting time' vs real pain is approval chase",
"metric windows defined differently per channel"
],
"discovery_confidence_note": "demand_signal=E, validated_demand=U",
"limitations": "Platform API access volatility unknown per channel",
"disposition": "prior_retained"
}
prior-mortgage_brokersrecord 13
{
"id": "prior-mortgage_brokers",
"industry": "mortgage_brokers",
"observed": "2026-08-27",
"evidence_class": "E/I",
"source": "research/actionmodel-builder-research-2026-08-26/phase-2/outputs/industry-atom-specifications.md#4.13",
"claim": "Mortgage broker demand centers on borrower qualification, condition chasing and file movement to close; lender portals are authoritative for status and conditions.",
"atoms": [
"intake_normalize",
"extract_structure",
"triage_route",
"follow_up_chase",
"report_digest"
],
"entities": [
"applicant",
"loan_application",
"lender",
"condition",
"document",
"property",
"milestone",
"broker",
"consent",
"exception"
],
"terminal_owner": "licensed_broker_or_loan_processor",
"precomputable": [
"application/condition state machines",
"LOS/lender-portal system split",
"document checklist archetypes per loan type",
"archetype fit (case/workflow + document chase)"
],
"must_ask": [
"LOS/CRM reality and lender panel",
"loan-type mix (drives checklist variants)",
"condition-aging pain points",
"compliance/consent posture for borrower data"
],
"contradiction_risks": [
"lender status conflicts with LOS state",
"'faster closes' wanted vs bottleneck is borrower document supply"
],
"discovery_confidence_note": "demand_signal=E, validated_demand=U",
"limitations": "Sensitive financial data; discovery is metadata/synthetic until authority proven",
"disposition": "prior_retained"
}
prior-property_managementrecord 14
{
"id": "prior-property_management",
"industry": "property_management",
"observed": "2026-08-27",
"evidence_class": "E/I",
"source": "research/actionmodel-builder-research-2026-08-26/phase-2/outputs/industry-atom-specifications.md#4.14",
"claim": "Property management demand centers on tenant inquiries, rent chasing, work orders and lease filing; emergency and legal paths must stay human-owned.",
"atoms": [
"intake_normalize",
"triage_route",
"follow_up_chase",
"sync_handoff",
"extract_structure"
],
"entities": [
"property",
"unit",
"tenant",
"lease",
"rent_charge",
"payment",
"work_order",
"vendor",
"inquiry",
"exception"
],
"terminal_owner": "property_manager_or_portfolio_ops_lead",
"precomputable": [
"inquiry/rent/work-order state machines",
"PMS + payment-ledger system model",
"vendor-dispatch actor model",
"archetype fit (intake desk + portal)",
"portfolio size inferable from public listings"
],
"must_ask": [
"PMS and payment reality; where leases actually live",
"work-order volume and vendor network",
"emergency escalation rules",
"rent-aging policy and legal thresholds"
],
"contradiction_risks": [
"payment pending vs paid ambiguity",
"'automate tenant comms' wanted vs legal notice constraints"
],
"discovery_confidence_note": "demand_signal=E, validated_demand=U",
"limitations": "Jurisdictional tenancy law variation unknown",
"disposition": "prior_retained"
}
prior-real_estaterecord 15
{
"id": "prior-real_estate",
"industry": "real_estate",
"observed": "2026-08-27",
"evidence_class": "E/I",
"source": "research/actionmodel-builder-research-2026-08-26/phase-2/outputs/industry-atom-specifications.md#4.15",
"claim": "Real estate demand centers on portal-lead response, viewing bookings and listing paperwork; speed-to-lead is the economic driver.",
"atoms": [
"intake_normalize",
"follow_up_chase",
"schedule_coordinate",
"extract_structure",
"report_digest"
],
"entities": [
"lead",
"contact",
"property",
"listing",
"viewing",
"agent",
"document",
"consent",
"status",
"exception"
],
"terminal_owner": "listing_agent_or_brokerage_ops_lead",
"precomputable": [
"lead/viewing/listing state machines",
"portal+CRM+calendar system model",
"public listing inventory is directly observable",
"archetype fit (CRM + scheduling)"
],
"must_ask": [
"CRM/portal mix and lead routing rules",
"team structure (solo agent vs brokerage)",
"document/compliance checklist per listing",
"response-time expectations"
],
"contradiction_risks": [
"stale listings visible publicly vs claimed 'all current'",
"lead-response claims vs measurable portal response times"
],
"discovery_confidence_note": "demand_signal=E, validated_demand=U",
"limitations": "Portal API access varies by market",
"disposition": "prior_retained"
}
prior-recruiting_staffingrecord 16
{
"id": "prior-recruiting_staffing",
"industry": "recruiting_staffing",
"observed": "2026-08-27",
"evidence_class": "E/I",
"source": "research/actionmodel-builder-research-2026-08-26/phase-2/outputs/industry-atom-specifications.md#4.16",
"claim": "Recruiting demand centers on candidate screening, interview scheduling and role filling; fairness-sensitive attributes and employment decisions bound automation.",
"atoms": [
"extract_structure",
"triage_route",
"schedule_coordinate",
"follow_up_chase",
"report_digest"
],
"entities": [
"candidate",
"role",
"application",
"resume",
"recruiter",
"interview",
"availability",
"consent",
"stage",
"exception"
],
"terminal_owner": "recruiter_or_hiring_manager",
"precomputable": [
"pipeline stage machines",
"ATS-centric system model",
"fairness boundary (no automated ranking decisions)",
"archetype fit (pipeline CRM + scheduling)",
"public job postings reveal role mix and volume"
],
"must_ask": [
"ATS/CRM reality and calendar integration",
"role types and volume (perm/contract/exec)",
"screening criteria ownership",
"candidate-consent and retention posture"
],
"contradiction_risks": [
"'faster screening' wanted vs fairness/consent constraints",
"stage definitions vary per client"
],
"discovery_confidence_note": "demand_signal=E, validated_demand=U",
"limitations": "Fairness-sensitive; discovery must exclude protected attributes",
"disposition": "prior_retained"
}
prior-saasrecord 17
{
"id": "prior-saas",
"industry": "saas",
"observed": "2026-08-27",
"evidence_class": "E/I",
"source": "research/actionmodel-builder-research-2026-08-26/phase-2/outputs/industry-atom-specifications.md#4.17",
"claim": "SaaS demand centers on MQL-to-demo conversion, churn reduction and shipping speed; authority is split across revenue, success and engineering owners.",
"atoms": [
"triage_route",
"follow_up_chase",
"monitor_alert",
"sync_handoff",
"report_digest"
],
"entities": [
"account",
"contact",
"opportunity",
"subscription",
"usage_event",
"support_ticket",
"release",
"incident",
"owner",
"exception"
],
"terminal_owner": "revops_customer_success_or_engineering_owner_by_exception",
"precomputable": [
"split-authority model (CRM/billing vs analytics vs CI)",
"pipeline/renewal/release state machines",
"public site + docs + status page reveal stack and maturity",
"archetype fit (ops desk spanning three planes)"
],
"must_ask": [
"CRM/billing/analytics/CI tool reality and metric definitions",
"which of the three planes (revenue, success, shipping) hurts most",
"tenant/data boundaries for usage data",
"team topology and exception ownership"
],
"contradiction_risks": [
"churn 'signal' definitions conflict between tools",
"engineering vs revenue priorities on same data"
],
"discovery_confidence_note": "demand_signal=E, validated_demand=U",
"limitations": "SaaS is also the most self-served industry; willingness to buy semi-custom unknown",
"disposition": "prior_retained"
}