Merge pull request 'Add replaceable image and video capability contracts' (#358) from codex/household-scenario-audit into main
CI / test (push) Successful in 1m45s

Merge PR #358
This commit was merged in pull request #358.
This commit is contained in:
m
2026-08-25 13:05:36 +00:00
2 changed files with 11 additions and 3 deletions
+8 -3
View File
@@ -6,9 +6,12 @@ package capability
type MultimodalOperation string
const (
MultimodalTranscribe MultimodalOperation = "TRANSCRIBE_AUDIO"
MultimodalUnderstand MultimodalOperation = "UNDERSTAND_IMAGE"
MultimodalRecognizeOCR MultimodalOperation = "RECOGNIZE_OCR"
MultimodalTranscribe MultimodalOperation = "TRANSCRIBE_AUDIO"
MultimodalUnderstand MultimodalOperation = "UNDERSTAND_IMAGE"
MultimodalRecognizeOCR MultimodalOperation = "RECOGNIZE_OCR"
MultimodalUnderstandVideo MultimodalOperation = "UNDERSTAND_VIDEO"
MultimodalGenerateImage MultimodalOperation = "GENERATE_IMAGE"
MultimodalGenerateVideo MultimodalOperation = "GENERATE_VIDEO"
)
type MultimodalResponse struct {
@@ -17,4 +20,6 @@ type MultimodalResponse struct {
Provider string `json:"provider,omitempty"`
ObservedAt int64 `json:"observed_at_unix_millis"`
ConfidenceNote string `json:"confidence_note,omitempty"`
ContentType string `json:"content_type,omitempty"`
DataBase64 string `json:"data_base64,omitempty"`
}
+3
View File
@@ -32,6 +32,9 @@ const (
CapabilityKindSpeechToText CapabilityKind = "SPEECH_TO_TEXT"
CapabilityKindImageUnderstanding CapabilityKind = "IMAGE_UNDERSTANDING"
CapabilityKindOCR CapabilityKind = "OCR"
CapabilityKindVideoUnderstanding CapabilityKind = "VIDEO_UNDERSTANDING"
CapabilityKindImageGeneration CapabilityKind = "IMAGE_GENERATION"
CapabilityKindVideoGeneration CapabilityKind = "VIDEO_GENERATION"
)
// Binding is Capability-owned correctness state. SecretRef is an opaque