Skills
skills/klich3/rocket-store-lcoal-memories-python/rocket-store/Gen Agent Trust Hub

rocket-store

Pass

Audited by Gen Agent Trust Hub on Mar 9, 2026

Risk Level: SAFEPROMPT_INJECTIONDATA_EXFILTRATIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes and stores external data without sanitization, creating a surface for indirect prompt injection.\n
  • Ingestion points: The scripts/store_data.py script accepts data_json directly from command-line arguments.\n
  • Boundary markers: Absent. No delimiters or warnings are used to distinguish stored data from agent instructions.\n
  • Capability inventory: The skill uses Rocketstore to perform file read and write operations on the local filesystem in scripts/store_data.py.\n
  • Sanitization: Absent. Data is parsed via json.loads and stored without content validation.\n- [DATA_EXPOSURE]: The skill lacks sanitization for collection and key parameters in scripts/store_data.py, which could potentially be exploited for directory traversal to read or write files outside the intended storage area.\n- [EXTERNAL_DOWNLOADS]: The skill depends on the Rocket-Store package from PyPI, which is a standard utility for file-based JSON storage.\n- [PERSISTENCE_MECHANISMS]: The skill is designed to create persistent JSON records on the local filesystem, which is its primary stated purpose.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 9, 2026, 11:20 AM