{"skill":{"slug":"chemical-recognition","displayName":"αExtractor","summary":"Optical chemical structure recognition workflow for extracting molecule structures and names from images through SciMiner.","tags":{"latest":"1.0.2"},"stats":{"comments":0,"downloads":271,"installsAllTime":0,"installsCurrent":0,"stars":0,"versions":3},"createdAt":1776528348328,"updatedAt":1778663855491},"latestVersion":{"version":"1.0.2","createdAt":1778423294328,"changelog":"- All agent payload construction is now strictly based on the internal script registry (`scripts/sciminer_registry.py`), replacing ad hoc or OpenAPI-based specification.\n- Invocation and parameter validation require use of `build_payload_from_registry`; agents must check all keys against the authoritative registry.\n- Updated documentation with exact examples for importing and using the registry for payload generation, including guidance on file upload and payload building.\n- Agents are explicitly instructed to never invent or guess parameter keys and to cite the registry in all summaries.\n- No workflow or API changes; all behavioral and implementation guidance for safer and more robust tool integration.","license":"MIT-0"},"metadata":null,"owner":{"handle":"sciminer","userId":"s170qp1q4twz35wa85ppa8894h83w461","displayName":"SciMiner","image":null},"moderation":null}