molcompose-mcp
Download statistics: Totals (Pepy). These are package downloads, not unique users or MCP calls.
An MCP stdio server for protein-interface analysis and molecular visualization in UCSF ChimeraX. Agents use the same MolCompose commands as the graphical panel.
Installation
Requires Python 3.11+, ChimeraX 1.12 and the MolCompose bundle. Install the latest published bridge:
python -m pip install --upgrade molcompose-mcp
This README describes MCP 0.1.2 with bundle 0.1.3, not yet published.
For this checkout, run python -m pip install ./mcp from the repository root
and install the matching bundle using the
checkout instructions.
In the MolCompose Agent tab
Open Tools → Structure Analysis → MolCompose, then select the Agent tab.
Select an installed, signed-in Codex or Claude Code CLI, then ask a question.
The tab starts the local bridge when needed and selects --profile assistant
automatically.
External MCP client
Start the REST bridge in ChimeraX:
remotecontrol rest start port 3000 json true
Use the actual port printed by ChimeraX. Register this command in your MCP client:
molcompose-mcp --chimerax-url http://127.0.0.1:3000 --profile assistant
{
"mcpServers": {
"molcompose": {
"command": "molcompose-mcp",
"args": [
"--chimerax-url", "http://127.0.0.1:3000",
"--profile", "assistant"
]
}
}
}
Use the executable's absolute path if your client cannot find it. To generate configuration with the resolved path, run:
molcompose-mcp --chimerax-url http://127.0.0.1:3000 --profile assistant --print-config
Copyable prompts: Connection setup · Verify analysis and preview.
Assistant workflow
| Tool | Use |
|---|---|
open_structure |
Open a supported coordinate file or PDB ID |
inspect_session |
Inspect models, chains and version compatibility first |
analyse_interface |
Analyse a chain pair or request clarification |
load_external_evidence |
Import ΔΔG, energy or RMSF data |
compose_figure |
Apply a preset for the requested view |
render_preview |
Inspect the image before exporting |
export_artifact |
Save the figure, recipe and provenance, with overwrite confirmation |
For structured results, continue only after status: completed; resolve
needs_input, needs_confirmation or failed before retrying. render_preview
returns an image instead, which requires visual inspection. Successful export
alone does not establish figure quality.
For advanced integrations, use --profile expert (29 tools). Omitting
--profile selects all (34 tools). The assistant profile has 7 tools.
See the MCP reference
for individual tools, result fields and safety restrictions.
Importing energy data
Both load_energy and load_external_evidence accept solvation: "pb" or
solvation: "gb". A file containing both methods requires a choice:
{"kind": "energy", "path": "/data/decomp.dat", "model": "#1", "chains": "A:A,B:B", "solvation": "pb"}
This example is for load_external_evidence; omit kind for load_energy.
Match the chain mapping to your structure. Imported energies are not recalculated.
Troubleshooting
- Connection: keep ChimeraX open and use its actual REST port. Call
inspect_sessionto check bundle/bridge compatibility. - Preview or export: use windowed ChimeraX on macOS, not
--nogui. - Missing metric: check
skippedand the required structure or input files. - Overwrite or upload: assistant tools ask for confirmation. For PythiaStudio,
configure
PYTHIASTUDIO_API_KEY; local imports are not uploaded. - Local access: the REST bridge is unauthenticated. Stop it with
remotecontrol rest stopwhen finished. MCP input validation is not a host sandbox.
Metadata
Release files for molcompose-mcp 0.1.2
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 | |
|---|---|---|---|---|
| molcompose_mcp-0.1.2.tar.gz | 92.4 kB | With provenance | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| molcompose_mcp-0.1.2-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 153.8 kB
Release files / molcompose_mcp-0.1.2.tar.gz
| Download URL | molcompose_mcp-0.1.2.tar.gz |
|---|---|
| Size | 92.4 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
90735d9267eb638fbb1cdc099a20069e72d0b39cba688d8034e8a11925c5a59f
|
|
BLAKE2b-256 checksum How to use checksums |
3ec9e581083ad32f1d3ac39763a66654de27b19d134abfa2b5ac8d4bd399a5d2
|
| 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 Sep 8, 2026.
Transparency logRelease files / molcompose_mcp-0.1.2-py3-none-any.whl
| Download URL | molcompose_mcp-0.1.2-py3-none-any.whl |
|---|---|
| Size | 61.5 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
0a3c8eb8ec958bb15343b2c8ee366a5f053b91457c67e0024cd98b8c5f7cffc8
|
|
BLAKE2b-256 checksum How to use checksums |
167e3dc994e389bf660c77bdbe83e833ac0ba93df251b76af28b09c8e77186ce
|
| 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 Sep 8, 2026.
Transparency logSecurity for molcompose-mcp 0.1.2
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 molcompose_mcp-0.1.2.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 Sep 8, 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.
To learn more about protecting yourself from malicious actors, visit our package security documentation.