agentsclimarketplace
Sign in

Ase workflows

Skill jinzhezenggroup/computational-chemistry-agent-skills/atomistic-workflows/ase/ase-workflows

Route ASE atomistic workflow requests to task-specific subskills based on user intent. Use when the user asks for ASE-based static, relaxation, MD, or NEB workflows and you must apply consistent workflow controls independent of calculator backend.From its SKILL.md

Install
npx -y skills add jinzhezenggroup/computational-chemistry-agent-skills --skill ase-workflows

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

What its file declares

Copied from the file, not written here

The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

2.0 KB, 333 tokens by cl100k_base, as published. Nobody here has run it

ASE Workflows Router

Use this skill as the top-level ASE workflow orchestration layer.

Purpose

This skill routes requests to one task-specific workflow subskill path:

  • ase/ase-workflows/static
  • ase/ase-workflows/relax
  • ase/ase-workflows/md
  • ase/ase-workflows/neb

Scope

This router skill should:

  • classify user intent into one ASE workflow type
  • collect minimal task context (goal, convergence target, outputs)
  • delegate detailed workflow controls to selected subskill
  • request calculator backend via ase/ase-calculators router
  • enforce consistent output/handoff policy

This router skill should not:

  • hardcode backend-specific calculator parameters
  • execute or submit calculations directly

Routing rules

  1. single-point energy/force/stress -> ase/ase-workflows/static
  2. geometry optimization -> ase/ase-workflows/relax
  3. finite-temperature trajectory -> ase/ase-workflows/md
  4. reaction path / transition-state pathway -> ase/ase-workflows/neb

If intent is ambiguous, ask one focused clarification question.

Shared policy for all workflow subskills

  • separate workflow logic from backend adapter logic
  • expose assumptions and unresolved choices
  • return reproducible task layout and run script
  • if execution is requested, hand off to dpdisp-submit

Output from router

Provide:

  1. selected workflow subskill path
  2. why it was selected
  3. missing minimum inputs (if any)
  4. explicit next step (invoke selected subskill)

What ships with it: 1 file

704 B alongside SKILL.md

references/

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.