The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external genomics data and metadata files which could potentially contain malicious instructions if the source data is compromised.
Ingestion points: The skill reads external files using Read10X_h5, read.csv, and CreateChromatinAssay in examples/signac_workflow.R.
Boundary markers: No explicit delimiters or instructions are used to separate data from the agent's processing logic.
Capability inventory: The skill possesses the ability to write analysis results to the file system (saveRDS, pdf) and execute external bioinformatic tools.
Sanitization: Input metadata from files like singlecell.csv is used for cell labeling and filtering without explicit sanitization.
[EXTERNAL_DOWNLOADS]: The skill references and provides commands for downloading software from external sources.
The usage-guide.md file contains instructions to install the ArchR package from the GreenleafLab/ArchR GitHub repository and mentions the AMULET tool from UcarLab/AMULET. These are well-known academic tools in the bioinformatics community.
[COMMAND_EXECUTION]: The skill interacts with system-level executables for genomic analysis.
The R code in SKILL.md and examples/signac_workflow.R includes calls to the MACS3 peak caller (via macs2.path or pathToMacs2), which involves executing a command-line tool with user-provided arguments.

bio-atac-seq-single-cell-atac