skills/aircury/ai-framework/airsync/Gen Agent Trust Hub

airsync

Pass

Audited by Gen Agent Trust Hub on Apr 20, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill defines a memory retrieval system that processes data from a shared knowledge base. While this represents a surface for indirect prompt injection, the system includes design patterns like a gated review process (INBOX to PUBLISHED) and citation requirements to mitigate the risk of malicious content being treated as instructions. \n
  • Ingestion points: Memory content retrieved via memory_search and memory_get in SKILL.md. \n
  • Boundary markers: The skill encourages source citation but does not explicitly define delimiters for retrieved content in its prompt examples. \n
  • Capability inventory: The agent can propose, promote, and archive memories. \n
  • Sanitization: No explicit content sanitization or instruction filtering is described. \n- [DATA_EXFILTRATION]: The skill facilitates the storage of knowledge in a managed memory system via the memory_propose tool. This is the intended purpose of the skill and is governed by specific safety rules that explicitly prohibit the inclusion of credentials, tokens, or personal data.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 20, 2026, 09:19 PM
Security Audit — agent-trust-hub — airsync