The Agent Skills Directory

[SAFE]: The skill operates entirely on local data provided by the user and does not attempt to access sensitive system files or establish network connections. All scripts (Python and R) follow standard data science practices using well-known libraries like pandas, matplotlib, and the R meta/metafor packages.
[INDIRECT_PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection as it ingests untrusted external data files to generate reports, though the risk is minimal due to type-validation during processing.
Ingestion points: scripts/convert_data.py reads data from Forest plot_continuity_test data_1.txt. The R and Python analysis scripts (scripts/forest_continuous.R, scripts/forest_continuous.py) ingest CSV files specified via command-line arguments.
Boundary markers: No explicit prompt delimiters or instructions to ignore embedded commands are present in the workflow.
Capability inventory: The skill is limited to writing PNG images and CSV data tables to the local file system. It executes local scripts via Rscript and python but does not possess network exfiltration or arbitrary code execution capabilities.
Sanitization: The scripts perform basic data validation by converting input columns to numeric types using pd.to_numeric (Python) and as.numeric (R), which effectively filters out non-numeric malicious strings in the data fields.

meta-forest-continuous-plot