snaplii-agent-payments
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The installation process involves cloning a remote repository from GitHub (
https://github.com/Snaplii-Inc/agent-to-merchant-payments.git) to install the Snaplii CLI tool. - [COMMAND_EXECUTION]: The skill provides multiple Python code examples that utilize
subprocess.runto execute shell commands for purchasing gift cards and managing inventory via the Snaplii CLI. - [DYNAMIC_EXECUTION]: The Python examples demonstrate dynamic assembly of shell commands at runtime, incorporating variables such as item IDs, prices, and province codes into subprocess calls.
- [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection because it ingests untrusted data from an external API and has high-impact capabilities.
- Ingestion points: The agent retrieves data from the Snaplii API including merchant brand details, gift card categories, and purchase quotes (e.g., via
snaplii browseandsnaplii quote). - Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded commands within the fetched merchant data.
- Capability inventory: The skill allows the agent to perform financial transactions (
snaplii purchase) and access sensitive redemption codes and PINs (snaplii giftcard detail). - Sanitization: There is no mention of sanitization, filtering, or validation of the text returned by the Snaplii API before it is processed by the agent.
Audit Metadata