← Back to Skills Marketplace
Protein Mining
by
majiajia0529
· GitHub ↗
· v1.0.0
· MIT-0
243
Downloads
0
Stars
1
Active Installs
1
Versions
Install in OpenClaw
/install protein-mining
Description
自动化蛋白质挖掘完整流程，从文献调研到潜在蛋白确定，生成可发表的学术成果
Usage Guidance
This skill describes a legitimate, heavy bioinformatics pipeline but the metadata is incomplete: it declares only three binaries while the instructions require many more tools (CD-HIT, TrimAl, PAML, CAFE, AlphaFold/ColabFold, Foldseek, PyMOL, Python packages, etc.) and it mentions a 'protein-mining' CLI that is not provided. Before installing or running it: (1) confirm where the 'protein-mining' executable comes from and obtain a trustworthy install method or source code; (2) ensure all required bioinformatics binaries are installed from verified projects (not random URLs); (3) be aware that using AlphaFold/ColabFold or Google Colab may send sequence data to third-party services — avoid submitting sensitive sequences; (4) run the workflow in an isolated environment (container/VM) if you will allow the agent to invoke commands autonomously; (5) ask the author for an explicit install script, dependency list, and provenance for the CLI. If you cannot verify those, treat the skill as untrusted and do not let it run commands on sensitive systems or data.
Capability Analysis
Type: OpenClaw Skill
Name: protein-mining
Version: 1.0.0
The protein-mining skill bundle describes a legitimate bioinformatics workflow for automated protein discovery and analysis. It outlines a 7-module process involving literature review, sequence collection, HMM modeling, and phylogenetic analysis using standard scientific tools like MAFFT, HMMER, and IQ-TREE. There are no signs of malicious code, data exfiltration, or prompt injection; the instructions are strictly aligned with the stated purpose of computational biology research.
Capability Assessment
Purpose & Capability
The described pipeline (MSA, HMM, phylogeny, AlphaFold, Foldseek, etc.) matches the stated purpose, but the registry 'requires.bins' lists only mafft, hmmbuild, iqtree while the SKILL.md lists many additional dependencies (CD-HIT, TrimAl, PAML, CAFE, AlphaFold2/ColabFold, Foldseek, PyMOL, etc.). The usage examples also reference a 'protein-mining' CLI that is not provided or declared. This mismatch suggests the skill assumes external software/environment setup that is not documented in metadata.
Instruction Scope
The SKILL.md outlines expected runtime actions (literature search, downloading sequences from UniProt/NCBI, running alignment/HMM/phylogeny and AlphaFold), all of which are relevant to the task. It does not instruct reading unrelated local files or accessing secrets, but it implicitly requires network access to public databases and potentially third-party services (e.g., ColabFold/AlphaFold on remote GPU instances). There are no explicit instructions to transmit data to unknown endpoints, but using Colab/third-party services can result in data leaving the local environment.
Install Mechanism
There is no install spec (instruction-only skill), so nothing is written to disk by the skill itself. That reduces immediate supply-chain risk. However, because the skill expects many external tools and an undocumented 'protein-mining' CLI, the user must manually install/trust these binaries from outside sources.
Credentials
The skill requests no environment variables or credentials, which is appropriate. However, SKILL.md expects use of external services (AlphaFold/ColabFold) that often require accounts or tokens in practice; those are not declared. Also heavy compute/GPU access will be required for structure prediction but this is a resource, not a secret.
Persistence & Privilege
The skill is not always-enabled and is user-invocable; it does not request persistent system privileges or modify other skills. Autonomous invocation is permitted (platform default) but not combined with other high-risk flags.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install protein-mining - After installation, invoke the skill by name or use
/protein-mining - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
- Initial release of the protein-mining skill for automated, end-to-end protein mining workflows.
- Covers literature survey, sequence collection and QC, HMM profile construction, cross-species searches, structure-based search (Foldseek), phylogenetic analysis, and candidate prioritization.
- Each module generates publication-ready data files, visualizations, and summary reports.
- Requires key bioinformatics software tools including MAFFT, hmmbuild, and IQ-TREE.
- Workflow and technical requirements detailed for seamless academic application.
Metadata
Frequently Asked Questions
What is Protein Mining?
自动化蛋白质挖掘完整流程，从文献调研到潜在蛋白确定，生成可发表的学术成果. It is an AI Agent Skill for Claude Code / OpenClaw, with 243 downloads so far.
How do I install Protein Mining?
Run "/install protein-mining" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Protein Mining free?
Yes, Protein Mining is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Protein Mining support?
Protein Mining is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Protein Mining?
It is built and maintained by majiajia0529 (@majiajia0529); the current version is v1.0.0.
More Skills