The Agent Skills Directory

[SAFE]: The skill implements standard Bayesian ranking algorithms commonly used for 'Top Rated' lists. The logic is transparent and mathematically sound.
[COMMAND_EXECUTION]: The provided Python script scripts/bayesian_avg.py uses standard libraries (argparse, json) to perform calculations. It does not utilize dangerous functions such as eval(), exec(), or subprocess to run shell commands.
[DATA_EXFILTRATION]: No network operations (curl, wget, requests, etc.) or external data transmission patterns were found. The script only reads from a user-specified local JSON file.
[PROMPT_INJECTION]: The SKILL.md file contains instructional content for the AI agent that defines its role and tasks clearly without any attempts to bypass safety filters or override system constraints.
[EXTERNAL_DOWNLOADS]: No external dependencies, remote scripts, or third-party package installations are required or performed by this skill.
[CREDENTIALS_UNSAFE]: No hardcoded API keys, passwords, or sensitive file paths (e.g., .ssh, .aws/credentials) were detected in the scripts or documentation.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-provided item data for ranking.
Ingestion points: scripts/bayesian_avg.py ingests data via a JSON file provided to the --input argument.
Boundary markers: The script processes structured JSON data, which inherently separates metadata from execution logic.
Capability inventory: The script is limited to mathematical operations and standard output of results.
Sanitization: The script performs input validation on numerical values (e.g., checking for non-negative review counts) to ensure calculation integrity.

algo-rank-bayesian