Documentation
¶
Index ¶
Constants ¶
View Source
const ( OutcomeSelected = "selected" OutcomeExcluded = "excluded" OutcomeShadowed = "shadowed" OutcomeDelivered = "delivered" ProviderCategoryPerson = "person-package" ProviderCategoryCatalogue = "catalogue" ProviderCategoryRole = "role-provider" )
Variables ¶
This section is empty.
Functions ¶
This section is empty.
Types ¶
type ProviderReport ¶
type ProviderReport struct {
Source string `json:"source"`
Category string `json:"category"`
Scope string `json:"scope"`
Outcome string `json:"outcome"`
Reason string `json:"reason"`
Warning bool `json:"warning,omitempty"`
Skills int `json:"skills"`
ContextBytes int64 `json:"context_bytes"`
ApproximateTokens int64 `json:"approximate_tokens"`
SelectorReason string `json:"selector_reason,omitempty"`
}
ProviderReport retains one provider outcome and its canonical selected-skill byte budget, which stays deterministic across native and staged projection.
type Resolution ¶
type Resolution struct {
Request *schema.Request
Person *person.Person
Personalities []string
Boundaries []string
RolePurpose string
RoleBriefing string
OperatingBase string
OperatingAppendix string
Instructions []Selected
Skills []Selected
CompiledBodies []Selected
FavoriteColor string
Warnings []string
Decisions []Decision
Providers []ProviderReport
SourceIDs []string
Repositories []schema.RepositorySelection
}
Resolution is the full composition plan: what was selected, how it is delivered, and the trace built while those choices were made.
Click to show internal directories.
Click to hide internal directories.