The Agent Skills Directory

[CREDENTIALS_UNSAFE]: No hardcoded credentials, API keys, or sensitive file paths (such as SSH keys or environment files) were found in the templates or the scoring script.
[REMOTE_CODE_EXECUTION]: The Python scoring engine scripts/process.py uses only standard Python libraries (json, datetime, dataclasses) and does not contain any remote code download or execution patterns.
[COMMAND_EXECUTION]: No shell commands, subprocess calls, or system-level modifications were detected in the skill content.
[DATA_EXFILTRATION]: There are no network operations, HTTP requests, or external data transmission patterns present in the code or instructions.
[PROMPT_INJECTION]: The skill body consists of structured regulatory templates and compliance guidance. It contains no instructions that attempt to bypass AI safety filters or override agent behavior.
[OBFUSCATION]: All files are in plain text markdown or Python. No Base64, hex encoding, zero-width characters, or homoglyph attacks were identified.
[DYNAMIC_EXECUTION]: The risk calculation logic is static and does not use eval(), exec(), or unsafe deserialization methods.

ai-privacy-impact-template