The Agent Skills Directory

[PROMPT_INJECTION]: No patterns of prompt injection, role-play instructions, or attempts to bypass safety filters were detected. The instructions are purely functional and oriented towards providing relocation guidance.
[DATA_EXFILTRATION]: The skill does not access sensitive local file paths (such as SSH keys or AWS credentials) and contains no code for exfiltrating data to external servers.
[REMOTE_CODE_EXECUTION]: There are no patterns of remote code execution. The skill does not perform external downloads of scripts or packages, and the provided Python scripts in the scripts/ directory perform local logical operations without network access.
[OBFUSCATION]: No obfuscated content, encoded strings (e.g., Base64), or hidden characters were found in the instructions, metadata, or source code.
[COMMAND_EXECUTION]: The skill instructs the agent to run specific local Python scripts with arguments based on user input. These scripts (relocation-checklist.py and residency-check.py) use standard argument parsing and do not use dangerous functions like eval() or exec().
[CREDENTIALS_UNSAFE]: No hardcoded credentials, API keys, or secrets were found. The skill correctly instructs users to manage their own logins for government portals.
[INDIRECT_PROMPT_INJECTION]: While the skill processes user-supplied parameters for relocation scenarios, these are passed as command-line arguments to internal scripts that do not interpolate them into prompts or execute them as code. The capability surface is limited to informational guidance.
[PRIVILEGE_ESCALATION]: No commands requiring elevated privileges (like sudo) or modifications to system configurations were detected.
[PERSISTENCE_MECHANISMS]: No attempts to establish persistence on the host system (e.g., cron jobs, startup items) were found.
[DYNAMIC_EXECUTION]: The provided scripts perform static logic based on user input. There is no runtime generation or execution of code from untrusted sources.

israeli-relocation-abroad