Repository navigation
feat/workbench-model-media-capabilities - #4
Merged
Merged
Conversation
…log llm calls - ModelInput accepts images from declared resources, https URLs or data URLs - llm.listModels, generateSpeech, generateImage and decide mirror the Desktop Host SDK - Media and decision calls keep Host catalog defaults instead of DEFAULT_LLM_MODEL - llm guide documents output resources, HOST_CREDITS_EXHAUSTED and Jev questions Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Sync the template SDK and
llmauthoring guide with the Desktop Host SDK additions from AoE Desktop branchfeat/workbench-model-media-capabilities(OpenSpecadd-workbench-model-media-capabilities).ModelInput.images: up to 8 images from a declared directory resource, an https URL (downloaded by the Host) or a data URL; text-only models fail withLLM_IMAGE_INPUT_UNSUPPORTEDbefore any charge.llmoperations:listModels,generateSpeech,generateImage(generation and reference-image edits),decide(platform Jev decision model, defaultjev-latest).read-writedirectory resource and returned as logical references only.DEFAULT_LLM_MODELstill applies to text calls only; media and decision calls use Host catalog defaults.HOST_CREDITS_EXHAUSTED(not-executed, Floatboat shows its own credits prompt) and the 330 s default deadline for image generation.The Host side lands in AoE Desktop; older Hosts report the new operations as unavailable through
capabilities.list. Video generation is intentionally not included.Verification
node node_modules/typescript/bin/tsc -p tooling/tsconfig.sdk.jsonpnpm skills:check— 16 topics, Codex source and Claude entry agreepnpm test:tooling— 33/34 on the first run; the remainingruntime-packagecase failed on cleanmainas well because localnode_moduleslackedjavascript-obfuscator, and passed afterpnpm install --frozen-lockfile🤖 Generated with Claude Code