The Agent Skills Directory

[SAFE]: Standard Library Usage: The implementation uses only built-in Python modules such as hashlib, pathlib, and json, which avoids risks associated with unvetted third-party dependencies.
[SAFE]: Secure Deserialization: The pattern employs json.loads and json.dumps for cache storage. This is a secure approach that avoids the critical vulnerabilities found in unsafe formats like pickle or yaml.load.
[SAFE]: File System Integrity: File access is restricted to reading specified input files for hashing and writing/reading JSON files within a designated cache directory. The use of hex-encoded SHA-256 hashes as filenames prevents path traversal vulnerabilities in the cache storage logic.
[SAFE]: Robust Error Handling: The code includes checks for file existence and gracefully handles JSON decoding errors by treating them as cache misses, ensuring the agent remains stable even if cache files are corrupted.

content-hash-cache-pattern