atmee-product

Warn

Audited by Snyk on Nov 20, 2026

Risk Level: MEDIUM
Full Analysis

MEDIUM W009: Direct money access capability detected (payment gateways, crypto, banking).

  • Direct money access detected (medium risk: 0.30). No in-scope operation is established for generation, clips, or conversations. Those are service costs: paid per call or per clip, and the agent is not documented as moving money. Plans and top-ups are handled by the human in Studio → Billing, not by the agent. The borderline item is sponsoring. Line 171 says a sponsor pool can be set up "directly through the agent or the CLI", and lines 142-143 describe visitors talking on the creator's credits from that pool. Under recall-first, this is treated as a possible direct operation that debits the owner's prepaid credit balance. The documented control is sign-in ("signed-in is enough"), which the rule accepts as an authenticated session. That makes the route medium rather than high. The skill itself does not run the pool; it only routes to it.

MEDIUM W013: Attempt to modify system services in skill instructions.

  • Attempt to modify system services in skill instructions detected (low risk: 0.10). The skill exposes remote, state-changing Atmee operations (API key create/rotate/revoke, avatar updates, sponsor funding) through a declared MCP integration, but no explicit unrecoverable operational loss is established, so gate 2 fails and the tier is capped at low.

Issues (2)

W009
MEDIUM

Direct money access capability detected (payment gateways, crypto, banking).

W013
MEDIUM

Attempt to modify system services in skill instructions.

Audit Metadata
Risk Level
MEDIUM
Analyzed
Nov 20, 2026, 06:16 PM
Issues
2