The Agent Skills Directory

[SAFE]: The skill is a utility for processing research data. It modifies a local CSV file (papers/extraction_table.csv) by adding and populating standard 'risk-of-bias' columns based on existing row data. It explicitly states no network access is required and none was found in the code.
[COMMAND_EXECUTION]: The execution script scripts/run.py performs a safe directory traversal to locate the project root and import internal environment helper modules (tooling.review_artifacts). It does not invoke system shells or execute user-provided strings.
[PROMPT_INJECTION]: Instructions provided in SKILL.md are descriptive and focus on maintaining data consistency and auditing quality. No patterns related to prompt injection, safety filter bypass, or system prompt extraction were detected.

bias-assessor