lib.opportunity-spec
Provider-neutral opportunity scoring and selection contracts for autonomous-company candidates.
@lssm-tech/lib.opportunity-specbun add @lssm-tech/lib.opportunity-specManifest declarations describe the current workspace; they do not verify registry publication or runtime execution.
Package guide
Provider-neutral opportunity scoring and selection contracts for autonomous-company candidates.
This package closes the first canonical ContractSpec slice for the Autonomous Company Opportunity Engine. It defines reviewable candidate, scorecard, evidence, gate, validation, simulation, portfolio, and handoff contracts; it does not research markets, call providers, mutate CompanyOS/Builder state, or approve production implementation.
Scope
Use this package to model:
- autonomous-company opportunity candidates;
- core scorecards: AFS, BVS, GFS, CES, DMS, RTS, RCS, SFS, and TMS;
- diagnostic scores: QAS, DRS, TVS, VCS, and OTS;
- per-score and per-estimate evidence claims from hypothesis to live customer metrics;
- hard kill gates, downgrade gates, HITL load, GTM/CAC/budget assumptions, validation experiments, simulations, and portfolio decisions;
- branded refs to CompanyOS, Builder, ProductIntent, growth, finance, assurance, and orchestration surfaces.
Boundary
This package is contract-only. It must not contain provider credentials, network calls, filesystem mutation, live research, customer-facing sends, production workflow creation, or module imports. High opportunity scores are decision evidence, not implementation approval.
Public entrypoints
@lssm-tech/lib.opportunity-spec— root barrel@lssm-tech/lib.opportunity-spec/types— Zod schemas and inferred types@lssm-tech/lib.opportunity-spec/validation— semantic validation helpers@lssm-tech/lib.opportunity-spec/validation-rules— reusable validation rule primitives for score coverage and evidence ref checks@lssm-tech/lib.opportunity-spec/fixtures— safe synthetic fixtures
Evidence thresholds
customer_validatedrequires evidence level3or higher.pilot_validatedrequires evidence level4or higher.live_provenrequires evidence level5.
Distribution-first drafts (additive, experimental)
@lssm-tech/lib.opportunity-spec/discovery exports OpportunityDraftSchema, OpportunityWaveSchema, scoped evidence/metric contracts, the owning OpportunityDiscoveryWaveFeature, and validateOpportunityDraft / validateOpportunityWave. Drafts preserve unassessed fields and unknown customer metrics; they do not require a company, canonical scores, budgets or simulations.
Commercial roles, acquisition mechanisms, creative production, publishing, offers and retention modes are independent dimensions. Catalog cash/effort/time ranges are editorial planning priors, never measured CAC, churn or conversion. Evidence supports its particular product, claim, audience, geography and period; vendor offers establish category facts, not our willingness to pay or demand.
Use the repository's scripts/opportunity-wave.ts captured-input runner to emit the generation prompt, validate, screen and export another wave. The runtime's separate selfserve-first@1.0.0 profile produces research/experiment priorities, without granting implementation, campaign or spending authority. Promotion to OpportunityCandidate requires independently supplied mature requirements and canonical validation. Existing mature contracts remain unchanged.
Wave 1.1.0: sales-assisted profile, production assets, channel variety
OpportunityWaveV11Schema (schemaVersion: "1.1.0") sits beside the frozen 1.0.0 schema; validate either with validateOpportunityDiscoveryWave.
profileIdisselfserve-firstorsales-assisted(OPPORTUNITY_DISCOVERY_PROFILE_REQUIREMENTS).sales-assistedreplaces theselfserve_purchasegate withbounded_sales_effortand theselfserve_adoptiondimension withassisted_conversion.- A play may declare
salesMotion: AI preparation, a human role, a handoff boundary. It passesbounded_sales_effortonly with at most two calls and at most 90 planned human minutes per closed account; a play above the ceiling must fail the gate and is deferred. Underselfserve-firsta mandatory call cannot passselfserve_purchase.sales_led,ai_sdr_callsandwebinar_demoplays must declare a motion undersales-assisted. - A play may declare
productionAsset(paid UGC, AI UGC, founder content, product demo, commissioned creator, customer content). It supplies creative, never reach. AI UGC declares a synthetic person; paid or synthetic people are never presented as customers; creator amplification needs explicit ad authorization. - A wave of eight or more drafts keeps any primary method at or below 25% and spans at least four method families (
OPPORTUNITY_WAVE_V11_CHANNEL_VARIETY).
Package verification
bun run testbun run typecheckbun run lint:checkbun run build
Relationships
Arrows point from a package to its direct dependency. Counts are derived from captured workspace manifests.
2 declarations
@lssm-tech/lib.contracts-spec-coreworkspace:^
Read the package manual ↗zodcatalog:
Declared capabilities
Provider-neutral opportunity scoring and selection contracts for autonomous-company candidates.
Scripts
publish:pkgbun publish --tolerate-republish --ignore-scripts --verbosepublish:pkg:canarybun publish:pkg --tag canarybuildcontractspec-bun-build buildbuild:bundlecontractspec-bun-build transpilebuild:typescontractspec-bun-build typesdevcontractspec-bun-build devcleanrimraf dist .turbolintbun lint:fixlint:fixnode ../../../../scripts/biome.cjs check --write .lint:checknode ../../../../scripts/biome.cjs check .testbun testprebuildcontractspec-bun-build prebuildtypechecktsgo --noEmit