메뉴
Use when the user asks to create, build, inspect, validate, or run computational workflows, DAG pipelines, relaxation→static chains, NEB/IRC workflows, or HPC job submission flows.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
| name | workflow |
| description | Use when the user asks to create, build, inspect, validate, or run computational workflows, DAG pipelines, relaxation→static chains, NEB/IRC workflows, or HPC job submission flows. |
All MCP workflow operations use the single catgo_workflow tool with an action parameter.
Do not assume old workflow docs are still correct.
Important current behavior:
create creates a workflow that already contains one structure_input node.run starts execution immediately. It does not open a confirmation dialog.connect defaults both handles to structure if omitted.node_types is only a discovery list. It does not give full node schema.| Action | Purpose | Required Params |
|---|---|---|
list | List all workflows | — |
templates | List available templates | — |
node_types | List node types by category | category? |
create | Create workflow | name, template_id? |
get | Read workflow graph | workflow_id |
add_node | Add node | workflow_id, node_type, params? |
remove_node | Remove node | workflow_id, node_id |
connect | Connect nodes | workflow_id, from_id, to_id, from_handle?, to_handle? |
set_params | Update params | workflow_id, node_id, params |
validate | Validate graph | workflow_id |
run | Start execution immediately | workflow_id, run_config? |
pause | Pause workflow | workflow_id |
resume | Resume workflow | workflow_id, run_config? |
status | Get run status | workflow_id |
step_error | Get failure detail | workflow_id, step_id |
node_types first for coarse discovery.templates if the user wants a preset.{"action": "create", "name": "TiO2 slab workflow"}
Immediately after create:
getstructure_input unless the workflow truly needs multiple independent inputs{"action": "add_node", "workflow_id": "...", "node_type": "geo_opt"}
{"action": "add_node", "workflow_id": "...", "node_type": "single_point"}
{"action": "set_params", "workflow_id": "...", "node_id": "...", "params": {"software": "vasp", "ENCUT": 520}}
Only omit handles when both sides are obviously single-structure nodes.
Safer form:
{
"action": "connect",
"workflow_id": "...",
"from_id": "n1",
"to_id": "n2",
"from_handle": "structure",
"to_handle": "structure"
}
create returns an empty graph.structure_input after create.connect can guess the right handle for multi-port nodes.validate warnings as harmless.run before the user has provided a valid run configuration when HPC settings matter.node_typescreategetadd_nodeset_paramsconnect with explicit handles when neededvalidategetrunCreate and manage computational chemistry workflows with CatGo. Supports VASP, CP2K, ORCA, MLP, LAMMPS. Build OER/HER/CO2RR workflows, geometry optimization, frequency analysis, Gibbs energy calculations.
Run and resume the CatGo md-orchestration poll loop — delegate each poll to a subagent (keep main context lean), verify convergence by force, auto-advance each converged species per-species (pipeline, not barrier), and resume a campaign from disk after context compaction / new session. Use when driving or resuming a campaign's job-watch loop. Pairs with catgo-campaign.
Run a CatGo file-first md-orchestration "campaign" — a multi-step / high-throughput computational-materials study (e.g. SAA HER screening) driven from a human-readable folder + markdown tree, not the DB workflow engine. Use when the user says "跑一个 campaign", "md 模式跑", "high-throughput screening", or wants an agent-in-the-loop study with stages/funnel/analysis/report. Requires `catgo` on PATH.
Route computational chemistry requests to the correct software and task skill. Entry point for all CatGo agent interactions.
Drive a file-first, agent-in-the-loop computational campaign via a folder + markdown tree (no DB). Use when the user opts out of the visual workflow engine.
Authoring conventions for CatGo md-orchestration campaigns — progressive markdown, README+INDEX pairs and keeping them current, logging interventions to LESSONS, human-readable (never-hash) names, and the progressive (top→stage→calc) plan. Use when creating/editing any campaign markdown (plan/README/INDEX/STATUS/LESSONS) so the file tree stays navigable and resumable. Pairs with catgo-campaign.