# Molclaw Pack Sidechains

> Predicts full-atom sidechain conformations from backbone PDBs using AttnPacker for structure preparation workflows.

- **Type:** Skill
- **Install:** `agentstack add skill-internscience-molclaw-molclaw-pack-sidechains`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [InternScience](https://agentstack.voostack.com/s/internscience)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [InternScience](https://github.com/InternScience)
- **Source:** https://github.com/InternScience/MolClaw/tree/main/skills/L1_tools/molclaw-pack-sidechains

## Install

```sh
agentstack add skill-internscience-molclaw-molclaw-pack-sidechains
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# AttnPacker Sidechain Packing

Note: 
- Local files are not directly accessible by the server. Please upload them to the server using `molclaw-file-transfer` before execution. 
- For PDB file inputs, it is recommended to preprocess them using `molclaw-pdbfixer` before execution.
- Please refer to skill `molclaw-scp-server` to complete tool invocation.

## Usage

### 1. Protein Sidechain Packing
The description of tool *pack_sidechains*.

```tex
Predict full-atom sidechain conformations from backbone PDBs for protein structure preparation workflows.
Args:
  input_pdb (str): Input PDB file path, required.
  device (str|None): Compute device such as cuda:0, default None (auto by source script).
  chunk_size (int): Inference chunk size for long proteins, default 500.
  no_post_process (bool): Skip rotamer post-processing for faster runtime, default False.
  max_optim_iters (int): Maximum optimization iterations in post-process, default 250.
  steric_wt (float): Steric clash penalty weight, default 1.0.
  optim_repeats (int): Post-process optimization repeats, default 2.
  dry_run (bool): Create a traceable run directory without running inference, default False.
Return:
  status (str): 'success', 'error', or 'partial_success'.
  msg (str): Human-readable execution message.
  input_pdb (str): Input PDB path used for this run.
  output_dir (str): Unique run directory under tool_result/pack_sidechains_result.
  output_pdb (str): Expected or generated output PDB path.
  device (str|None): Device value used for execution.
  chunk_size (int): Chunk size used.
  no_post_process (bool): Whether post-process was skipped.
  max_optim_iters (int): Max optimization iterations used.
  steric_wt (float): Steric weight used.
  optim_repeats (int): Optimization repeats used.
  dry_run (bool): Whether dry-run mode was used.
  error_type (str, optional): Exception type when status is 'error'.
  traceback (str, optional): Python traceback when status is 'error'.
```

How to use tool *pack_sidechains* :

```python
response = await client.session.call_tool(
    "pack_sidechains",
    arguments={
        "input_pdb": "/path/to/input.pdb",
        "device": "cuda:0",
        "chunk_size": 500,
        "no_post_process": False,
        "max_optim_iters": 250,
        "steric_wt": 1.0,
        "optim_repeats": 2,
        "dry_run": False
    }
)
result = client.parse_result(response)
output_pdb = result["output_pdb"]

```

#### Example parameter sets

```python
# 1) Main mode
{
    "input_pdb": "/path/to/input.pdb",
    "device": "cuda:0",
    "chunk_size": 500,
    "no_post_process": False,
    "max_optim_iters": 250,
    "steric_wt": 1.0,
    "optim_repeats": 2,
    "dry_run": False
}

# 2) Variant mode
{
    "input_pdb": "relative/path/to/test_backbone.pdb",
    "chunk_size": 500,
    "dry_run": True
}
```

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [InternScience](https://github.com/InternScience)
- **Source:** [InternScience/MolClaw](https://github.com/InternScience/MolClaw)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-internscience-molclaw-molclaw-pack-sidechains
- Seller: https://agentstack.voostack.com/s/internscience
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.