DFT Agent
Version 0.4.4. 中文说明
Qirui Cui · KTH Royal Institute of Technology
Describe a calculation, review its inputs, and run VASP or Quantum ESPRESSO on your own HPC cluster. Each project keeps its files, conversations and results together. You can revise tasks in chat, combine both engines in a workflow, and discuss the results when they return.
Conceptual overview. Current engines: VASP and Quantum ESPRESSO.
The built-in workflows cover structure edits, relaxation, SCF, bands, DOS, magnetism, SOC, DFT+U and hybrid functionals. QE also supports installation, phonons, NEB and electron–phonon calculations.
Install
Download v0.4.4:
- Apple Silicon Mac, macOS 14+: open the
.dmg, copy DFT Agent to Applications, then open it. - Windows x64: run the
.exeinstaller, then open DFT Agent from the Start menu.
Python and model SDKs are included. Codex CLI and Claude Code CLI are optional, separate installations. Linux users can install from source.
Start a project
- Click New project and give it a name.
- Open Connections → Model. Connect a signed-in Codex CLI or Claude Code CLI, or enter your provider's API key and model name. Test connection checks a saved connection. Model setup.
- Open Cluster and enter your SSH address, such as
ssh name@cluster. The app tries your keys first, then asks for a password or login code when needed. Review the detected software, or add it under Software. If QE is missing, review and start its installation here. Cluster setup. - Return to chat. Open Context → Files → Add files, upload your structure, and click Import files. Describe what you want to calculate and refine the plan in the same conversation.
- Click Prepare inputs, review the files and resources, tick the confirmation box, and click Submit. Follow the task in Context → Jobs and open its outputs under Results.
VASP and QE are selected per workflow step. A project can use either engine or both; it does not need a separate global switch. Attach results selects the runs for a follow-up discussion.
Model calls use your own account and quota. API keys and saved SSH passwords stay in the system keychain; login codes are not saved. VASP requires licensed VASP and POTCAR files. QE uses public pseudopotentials. No model credits or cluster allocation are included.
Explore offline
Open Settings → Legacy workbench → Runs → Load sample runs to browse the bundled SrTiO3, Si, MgO and Fe results. The legacy workbench also retains manual input preparation, the original forms and Memory browser. Use Settings → Language / 语言 to change the project interface language.
Quickstart · Project guide · Workflows · Scientific guidance · Memory
Citation
Cui, Q. (2026). DFT Agent (v0.4.4). Source.
Install · Windows · Methods · Input defaults · CLI examples · VASP validation · QE validation · MIT license · Notices · Citation
1 maintainer
Metadata
1 maintainer
Release files for dft-agent 0.4.4
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | |
|---|---|---|---|---|
| dft_agent-0.4.4.tar.gz | 21.6 MB | With provenance | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| dft_agent-0.4.4-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 27.4 MB
Release files / dft_agent-0.4.4.tar.gz
| Download URL | dft_agent-0.4.4.tar.gz |
|---|---|
| Size | 21.6 MB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
a4b2c8768d57ff5b906435fb7870b813c1b03cf7bdb50aef3c6656f27e95a980
|
|
BLAKE2b-256 checksum How to use checksums |
155b7d8f89ef0326d038ee378d96dc4aa0dbf503ce897abcaf7ec5cbd96b28cd
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 20, 2026.
Transparency logRelease files / dft_agent-0.4.4-py3-none-any.whl
| Download URL | dft_agent-0.4.4-py3-none-any.whl |
|---|---|
| Size | 5.7 MB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
9d176ebd2631a16b82d71dca6013a0b07a4eead03dbb1c7aae0a4362f2e03bbc
|
|
BLAKE2b-256 checksum How to use checksums |
942c267bb6c9367d3c4b660cd4e9c25c3f8b321f1b07d3cb0c9f4d6bb0d6576c
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 20, 2026.
Transparency logSecurity for dft-agent 0.4.4
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
Full provenance
All 2 files in this release have been uploaded with provenance metadata.
Attestation for dft_agent-0.4.4.tar.gz
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 20, 2026.
Transparency logA note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
