← Back to Skills Marketplace
266
Downloads
1
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install wemol-cli
Description
Use when a task may involve Wemol, a drug-discovery computing platform that integrates many biology, AI, and chemistry modules for workflows such as biologic...
Usage Guidance
This skill is a detailed manual for the wemol-cli and otherwise coherent, but be cautious about installing or running the CLI exactly as the docs show: the installer uses 'curl | sh' and PowerShell 'iex' against wemol.wecomput.com, and binaries are currently unsigned on macOS/Windows. Before installing, verify the vendor domain and inspect the installer script (do not blindly pipe it to a shell). Avoid changing the release base URL unless you fully trust the alternative host. Treat any prompts for username/password as requests for your Wemol account credentials only; do not share unrelated secrets. If you need stronger assurance, ask the skill author for a provenance link (release repository, GitHub/Git tags, or published checksums) or prefer an installer distributed via a well-known package host or signed releases.
Capability Analysis
Type: OpenClaw Skill
Name: wemol-cli
Version: 1.0.0
The wemol-cli skill bundle is a comprehensive and well-documented set of instructions for an AI agent to operate the Wemol drug-discovery platform. It includes detailed workflows for installation, authentication, and job management, with specific 'Hard Rules' and 'Detection Triggers' (SKILL.md) to ensure the agent uses the CLI reliably and avoids common errors. While it utilizes high-risk patterns like shell-based installers (install.sh from wemol.wecomput.com) and credential handling, these are strictly aligned with the tool's stated purpose and lack any evidence of malicious intent or deceptive prompt injection.
Capability Assessment
Purpose & Capability
Name/description and all included examples and reference files consistently describe a CLI client for a drug-discovery platform (discover modules/flows, submit jobs, download outputs). Nothing in the docs asks for unrelated cloud credentials or unrelated system access.
Instruction Scope
The SKILL.md stays on-topic: it tells the agent which wemol-cli commands to run and how to interpret outputs. It does reference reading a repository file (crates/cli/Cargo.toml) only when 'working inside this repo' and mentions environment overrides (WEMOL_HOST, WEMOL_SESSION_ID). It also documents non-interactive login (username/password) which could cause the agent to request/store user credentials if used; that behavior is expected for a CLI but should be explicit to users.
Install Mechanism
Although there is no compiled code bundled in the skill, the install instructions recommend piping a remote installer into sh or PowerShell (curl https://wemol.wecomput.com/... | sh and irm https://wemol.wecomput.com/... | iex). This is a high-risk install pattern (remote script execution) and the installer supports overriding the release base URL, which could be abused to point to a malicious artifact server. The domain appears vendor-specific (wemol.wecomput.com) rather than a major audited release host; that reduces but does not eliminate risk. The docs also warn binaries are unsigned on macOS/Windows, increasing user risk at install/run time.
Credentials
The skill declares no required env vars or credentials, which matches the doc being an operational manual. However the runtime instructions reference WEMOL_HOST, WEMOL_SESSION_ID and document non-interactive login with --username/--password; those are reasonable for a CLI but are effectively credentials the user may be asked to provide. The skill does not request unrelated secrets, but it could cause users/agents to supply Wemol account credentials.
Persistence & Privilege
The skill is not force-included (always:false) and requests no elevated platform privileges. The documented CLI installer writes into the user home (~/.wemol/bin or $HOME\.wemol\bin) and job downloads create .wemol-download-manifest.json — these are normal for a per-user CLI and consistent with the described purpose.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install wemol-cli - After installation, invoke the skill by name or use
/wemol-cli - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
wemol-cli-official 1.0.0
- Initial release of the official skill and operational manual for wemol-cli.
- Provides a comprehensive guide for discovering modules/flows, submitting jobs, tracking progress, and handling outputs using Wemol CLI.
- Includes practical command examples, core usage rules, and troubleshooting advice for biology, chemistry, and AI workflows.
- Covers authentication, session management, job inspection, result retrieval, and multilingual support.
- Emphasizes use of live command outputs and schemas over assumptions or static code edits.
Metadata
Frequently Asked Questions
What is wemol-cli?
Use when a task may involve Wemol, a drug-discovery computing platform that integrates many biology, AI, and chemistry modules for workflows such as biologic... It is an AI Agent Skill for Claude Code / OpenClaw, with 266 downloads so far.
How do I install wemol-cli?
Run "/install wemol-cli" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is wemol-cli free?
Yes, wemol-cli is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does wemol-cli support?
wemol-cli is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created wemol-cli?
It is built and maintained by wecomput (@nanomolar); the current version is v1.0.0.
More Skills