The Agent Skills Directory

Prompt Injection (SAFE): No instructions found that attempt to override AI behavior or bypass safety filters.
Data Exposure & Exfiltration (SAFE): No hardcoded credentials or access to sensitive file paths (e.g., SSH keys, AWS configs) were detected. Network operations are absent.
Obfuscation (SAFE): The content is clear and uses standard encoding. No hidden characters or multi-layer encoding found.
Unverifiable Dependencies & Remote Code Execution (SAFE): The Python script uses only standard libraries (os, sys, argparse, pathlib). There are no external package installations or remote code downloads.
Privilege Escalation (SAFE): No commands related to sudo, administrative permissions, or system configuration modifications were identified.
Persistence Mechanisms (SAFE): No attempts to modify startup scripts, cron jobs, or registry keys were found.
Indirect Prompt Injection (LOW): While the validation script reads external BED files, it does not execute their content or pass it to dangerous functions. The scope is limited to format validation.
Ingestion points: scripts/validate_files.py reads user-provided BED files.
Boundary markers: None.
Capability inventory: None. The script only performs print operations and file existence checks.
Sanitization: Basic validation of column counts and integer types is performed.

deeptools