The Agent Skills Directory

[PROMPT_INJECTION] (SAFE): The instructions are tailored for the stated purpose of data sourcing optimization and do not contain bypass or override patterns. Phrases such as "You are an expert" are used as part of the intended persona definition and do not attempt to bypass safety guidelines.
[DATA_EXFILTRATION] (SAFE): No hardcoded credentials, sensitive file access, or unauthorized network communication were found. Code templates use generic placeholders and mock functions like callProvider and checkCache.
[EXTERNAL_DOWNLOADS] (SAFE): The skill does not reference or download any external scripts or packages. All dependencies described are conceptual for integration design.
[COMMAND_EXECUTION] (SAFE): No dangerous command execution or subprocess calls are present. The provided Python script scripts/cost_calculator.py performs simple arithmetic calculations and volume discount logic.
[INDIRECT_PROMPT_INJECTION] (SAFE): While the skill is designed to handle data from external enrichment sources, it is presented as a framework and template library. It does not actively process untrusted external data within its own execution context in a way that would trigger a vulnerability.

data-sourcing