lca-hybrid-search

Plugin lca-hybrid-search from marketplace tiangong-lca-skills in tiangong-lca/agent-skills. Commands read from the repo's manifest — yesterday.

Skills (24)

• current-account-dataset-review

  Review current-account datasets and apply explicit, evidence-supported repairs through typed CLI operations. Use for inspecting owned drafts, references and changes before separate saves.

• dataset-rls-maintenance

  Orchestrate current-user RLS-scoped TianGong dataset cleanup, delete, retirement, and redo workflows through CLI maintenance plans, explicit approvals, and readback verification.

• embedding-ft

  Execute and troubleshoot Supabase edge function `embedding_ft` that consumes PGMQ embedding jobs, calls AWS SageMaker embeddings, and writes vectors back to Postgres. Use when validating job payload handling, investigating failed embeddings, tuning ack semantics, or adjusting worker auth/environment

• external-dataset-curated-import

  Guide external LCA packages through supported CLI conversion, evidence-led candidate review and separate atomic persistence, or assist an existing managed Foundry task without changing its owner protocol.

• flow-governance-review

  Inspect Flow identity, content and references using public rules and current CLI validation. Use for evidence-led Flow governance and explicit repair proposals without automatic merging or specialized repair execution.

• flow-hybrid-search

  Find and compare TianGong flow datasets through the installed CLI, select an explicit search scope and inspect exact candidate versions. Use for dataset discovery and choosing suitable modeling references.

• foundry-tidas-authoring

  Produce evidence-backed TIDAS decisions or patches for a concrete semantic work item generated by TianGong Foundry. Use only after the current task supplies its authoring package and contract context; return data files to the invoking workflow for validation and submission.

• foundry-tidas-import

  Use TianGong Foundry to import an external LCA dataset package or develop TIDAS data from source evidence, continue an existing Foundry task, resolve its semantic work items, and complete authorized writes with verified readback.

• lca-publish-executor

  Prepare validated datasets for explicitly authorized review submission and distinguish draft persistence, review and data product publication. Use when a user asks to submit or continue dataset review.

• lca-study

  Set up, develop or continue an LCA modeling study from a product goal, user data or an existing Model; guide evidence intake, foreground modeling, background selection, calculation, interpretation and interactive HTML/PDF research reporting. Also use for TianGong Study scenarios and revision-bound r

• lifecycleinventory-review

  Review canonical Process, Flow or Model data with current CLI validation and evidence-led scientific interpretation. Use for local inventory checks or reviewing explicitly selected remote versions.

• lifecyclemodel-automated-builder

  Develop an explicit Lifecycle Model from selected Process datasets and evidence using current CLI graph, validation and write operations. Use for agent-assisted model authoring; generation and provider decisions remain with the caller.

• lifecyclemodel-hybrid-search

  Find and compare TianGong model datasets through the installed CLI, select an explicit search scope and inspect exact candidate versions. Use for dataset discovery and choosing suitable modeling references.

• lifecyclemodel-recursive-orchestrator

  Organize evidence-led multi-level Model authoring from exact Process and Model references. Use for decomposing a modeling task and resolving explicit dependencies through current atomic CLI operations.

• lifecyclemodel-resulting-process-builder

  Generate a resulting Process candidate from an explicit Model using the current CLI aggregation operation, then validate it and perform a separate authorized create or save.

• process-automated-builder

  Author evidence-supported canonical Process datasets using current CLI contracts, validation and explicit writes; preserve scientific meaning and existing managed Foundry task boundaries.

• process-dedup-review

  Compare explicitly selected Process candidates and explain whether they represent the same scientific object. Use for duplicate review without automatic merging, deletion or reference replacement.

• process-hybrid-search

  Find and compare TianGong process datasets through the installed CLI, select an explicit search scope and inspect exact candidate versions. Use for dataset discovery and choosing suitable modeling references.

• process-scope-statistics

  Count and summarize a clearly selected Process dataset scope through current CLI statistics. Use for state or classification distributions and honest reporting of bounded scans.

• source-evidence-dataset-development

  Orchestrate evidence-driven TianGong TIDAS data creation or update from PDFs, Word files, web pages, URLs, APIs, databases, reports, or scientific literature.

• tiangong-lca-remote-ops

  Perform explicitly scoped remote dataset reads and authorized atomic draft writes with the current CLI. Use for exact readback, reference inspection and evidence-supported Process changes.

• tidas-bilingual-transcreation

  Produce evidence-supported translations for explicitly selected local TIDAS fields through typed CLI extract/apply operations, without changing scientific facts or saving remotely.

• tidas-contract-context

  Read the installed CLI's typed TIDAS schema, methodology and public-rule context before authoring or validating dataset content.

• tidas-data-import

  Select the appropriate path for importing structured LCA packages or authoring TIDAS data from source evidence through current typed CLI contracts and validation.