AI Agent Hub
Back to skills
Tencent tFold Antigen Antibody Complex Structure Prediction Model icon

Tencent tFold Antigen Antibody Complex Structure Prediction Model

Professional Updated 2026.08.30

Paste the following prompt into your AI chat to install this skill:

Please follow https://skillhub.cn/install/skillhub.md to install @tencent-adm/tfold-collection-skill.

About this skill

Problem

Nextflow omics apps such as tFold Collection (Nextflow) usually require checking login state, project configuration, sub-app AppId, Nextflow version, and import name before submission. Manual command assembly can accidentally use the collection AppId as a runnable app, or import a duplicate name and fail.

How It Works

  • Scoped collection: only runs the 4 sub-apps in tFold Collection (Nextflow), including tFold_ab_predict, tFold_ag_predict, tFold_ag_predict_icf, and tFold_tcr_predict.
  • Preflight checks: verifies omics CLI, whoami, and config show; C-end users also get quota checks before the first submission and before each run.
  • Execution guardrails: matches a target from the hardcoded sub-app list, blocks running the collection directly, performs duplicate-name checks, and asks for confirmation before omics run.
  • Failure handling: gives explicit guidance for DUPLICATE_APP_NAME, missing nf-version, and authentication failures, and reminds users of the outdir result path.

Boundaries

This skill targets public app runs on the Tencent Health Omics Platform. It does not cover other platforms, project-owned apps, or out-of-scope direct CLI usage. Local file inputs must be uploaded to the user's bound COS bucket first.

Use Cases

  • Select a tFold antibody, antigen, or TCR sub-app on Tencent Health Omics and submit a Nextflow structure prediction job.
  • Verify login, project, and COS bucket settings before submission while avoiding collection AppId misuse or duplicate imports.
  • Check remaining trial days and today's run quota before C-end users submit tFold jobs to prevent quota exhaustion.
  • Upload AI-generated or local FASTA files to the bound COS bucket first, then use them as inputs for tFold sub-apps.

Best For

  • Structural biology engineers: need to submit tFold antibody, antigen, ICF, or TCR prediction jobs to the omics platform accurately.
  • B-end Tencent Health Omics project members: need to verify project, environment, and COS bucket before running public Nextflow apps.
  • C-end trial users: want to run tFold predictions within trial quota and understand remaining days and daily runs.
  • Platform ops or app owners: need to manage submissions using fixed sub-app lists, duplicate checks, and run confirmation.