ClawHub

ClawBio Equity Scorer

v0.1.0

Compute HEIM diversity and equity metrics from VCF or ancestry data. Generates heterozygosity, FST, PCA plots, and a composite HEIM Equity Score with markdow...

⭐ 0· 286·0 current·0 all-time
by@manuelcorpas
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
✓
Purpose & Capability
Name/description match the included Python implementation and declared dependencies (numpy, pandas, scikit-learn, matplotlib, biopython). The required binaries (python3) and listed packages are appropriate for genotype parsing, PCA, plotting, and CSV handling. Minor mismatch: SKILL.md mentions Biopython VCF parsing but the visible code implements a simple line-based VCF parser rather than using Bio.SeqIO — this is an implementation choice, not a likely security issue.
✓
Instruction Scope
SKILL.md instructions confine the agent to detecting input type (VCF or CSV), parsing sample/population data, computing metrics, generating plots and a markdown report, and writing reproducibility artifacts (commands.sh, environment.yml, checksums). These actions are directly related to the stated purpose and do not direct the agent to read unrelated system files, environment variables, or to transmit data externally.
ℹ
Install Mechanism
Install spec lists standard Python packages (biopython, pandas, scikit-learn, matplotlib, numpy). The installer 'kind: uv' is uncommon in this manifest format — confirm what installer 'uv' maps to in your runtime (likely a PyPI/pip install). Packages themselves are well-known; no arbitrary URL downloads or archive extraction are present.
✓
Credentials
The skill requests no environment variables, no credentials, and no config paths. That is proportionate for an offline data-analysis tool that operates on user-supplied files.
✓
Persistence & Privilege
always:false (no forced persistent inclusion). The skill does not declare modifications to other skills or system-wide settings. It writes output files to an output directory (report, figures, environment export), which is normal for a reporting tool.
Assessment
This skill appears coherent with its stated purpose and contains only local analysis code and common Python dependencies. Before installing, consider: (1) Genomic data is highly sensitive — run the tool on a trusted, isolated machine or container and ensure input files do not leave your environment. (2) Confirm what 'uv' maps to in your runtime's installer and consider installing packages in a virtual environment or conda environment to avoid system-wide changes. (3) Review the full equity_scorer.py file (the provided excerpt is large but truncated in the report) to verify there are no hidden network calls or telemetry before running on confidential data. (4) Check resource needs for large VCFs (memory/CPU) and test on a small sample first. If you need, I can scan the remainder of the source for network I/O or other suspicious patterns.

Like a lobster shell, security has layers — review code before you run it.

latestvk972vesd64f2zqye41h851ewe581ttze

License

MIT-0
Free to use, modify, and redistribute. No attribution required.
Termshttps://spdx.org/licenses/MIT-0.html

Runtime requirements

⚖️ Clawdis
OSmacOS · Linux
Binspython3

Install

uvuv tool install biopython
uvuv tool install pandas
uvuv tool install scikit-learn
uvuv tool install matplotlib
uvuv tool install numpy

Comments