vibeflow-fds

name	vibeflow-fds
description	Use when preparing, running, debugging, post-processing, or extending VibeFlow/VibeCAE FDS fire, smoke, ventilation, combustion, sprinkler, evacuation-adjacent, and thermal hazard workflows from 0-caseDict/caseDict, including .fds input files, mesh blocks, devices, slices, boundary conditions, HRR output, Smokeview handoff, and reproducible fire-safety studies.

VibeFlow FDS

This skill is responsible for using FDS as VibeCAE's backend for fire dynamics, smoke spread, ventilation, thermal hazards, and fire-safety engineering.

• Usage mode: generate .fds input from caseDict, run FDS, and read .out, CSV, and Smokeview files.
• Development mode: develop FDS input templates, fire-source/material libraries, scene generators, post-processors, or Smokeview integration scripts.

When to Use

• The user task involves fire, smoke, heat release rate, ventilation, sprinklers, thermal radiation, visibility, temperature probes, or building-space safety assessment.
• .fds files need to be generated or repaired.
• _hrr.csv, _devc.csv, .smv, slice, boundary, particle, or isosurface outputs need to be read.
• Visualization handoff to Smokeview or ParaView is needed.

First Decide: Usage Mode or Development Mode

• Current scene modeling, execution, and post-processing: use usage mode.
• Template libraries, building-scene generators, material/combustion parameter libraries, or post-processing tools: use development mode.
• If FDS/Smokeview commands are unavailable or MPI issues occur, switch to vibeflow-toolchain first.
• If geometry comes from CAD or BIM and requires preprocessing, switch to the geometry/mesh-related skill first.

What to Read First

• The spatial dimensions, obstructions, vents, fire sources, materials, mesh, probes, output objectives, and safety metrics in 0-caseDict/caseDict.
• references/fds-playbook.md
• references/fds-practical-tips.md: software knowledge base, key concepts, failure interpretation, and adapter design points.
• Existing .fds files, material parameters, or experimental conditions in C-run/fds/ and Ref/.

Inputs

• 0-caseDict/caseDict
• Building/compartment dimensions, fire-source curves, material thermophysical properties, ventilation boundaries, and probe locations
• FDS input files or public benchmarks

Outputs

• C-run/fds/case-*/case.fds
• C-run/fds/case-*/Allrun / Allclean
• FDS logs, .smv, CSV, and slice/boundary/particle outputs
• D-post/fds_postprocess.py
• Charts for temperature, smoke layer, HRR, visibility, probes, and safety metrics

Recommended Structure

C-run/fds/
└── case-1/
    ├── case.fds
    ├── logs/
    ├── outputs/
    ├── Allrun
    └── Allclean
D-post/
└── fds_postprocess.py

Usage Mode Workflow

1. Map caseDict to FDS namelists: &HEAD, &MESH, &TIME, &REAC, &MATL, &SURF, &OBST, &VENT, &DEVC, &SLCF, &BNDF, and &TAIL.
2. Check units, mesh size, fire-source heat release rate, combustion model, and output frequency.
3. Generate case.fds, ensuring that mesh blocks, obstructions, and boundaries do not contradict each other.
4. Run a short smoke test to confirm input parsing, valid geometry, and output-file generation.
5. Run the production case and record .out, CSV, and visualization files.
6. Post-process HRR, device curves, temperature/velocity/smoke slices, and key safety metrics.
7. Write run status, physical assumptions, failure reasons, and output records back to caseDict.

Development Mode Workflow

1. Define the development target: input templates, room/building generators, material libraries, fire-source libraries, post-processing scripts, or Smokeview tasks.
2. Write development notes first, recording namelist interfaces, default parameters, validation scenarios, and risks.
3. All templates must explicitly preserve units and safety-related assumptions.
4. For post-processors, prefer reading CSV and structured outputs rather than relying on manual image inspection.
5. Validate parsing and short runs with at least one small room/simple fire source.

Core Rules

• Length units in FDS input are usually SI, and units in caseDict must be consistent.
• MESH resolution directly affects LES quality; do not optimize only for "runs quickly".
• Fire sources, materials, ventilation, and boundary conditions must be checked as a coupled set.
• Output frequency must balance evidence needs and file size.
• Smokeview is a visualization entry point, not a substitute for numerical post-processing evidence.

Validation

• The FDS executable is available.
• The .fds file parses successfully.
• A short run generates .out, CSV, or .smv.
• The post-processing script can read key outputs.
• Key safety metrics correspond to the user's objectives.

Fallback Rules

• Insufficient fire-source/material/ventilation information: return to vibeflow-case-dict
• FDS environment unavailable: return to vibeflow-toolchain
• Geometry input conflicts: return to the geometry stage
• Outputs insufficient to support reporting: return to the post-processing stage