phone-calls

Pass

Audited by Gen Agent Trust Hub on Apr 4, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of the agentcash CLI tool from the npm registry. This dependency is used to facilitate payments and API communication with the StablePhone service.\n- [REMOTE_CODE_EXECUTION]: The skill utilizes npx agentcash@latest to execute code directly from the npm registry. This allows the agent to perform actions using the latest version of the payment utility.\n- [COMMAND_EXECUTION]: The skill executes shell commands to interact with the agentcash CLI, including operations for checking balances (npx agentcash balance), managing account funding (npx agentcash accounts), and making network requests (npx agentcash fetch).\n- [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection because it processes transcripts and summaries of phone calls. Since call content is controlled by external parties, spoken instructions could potentially influence the agent's behavior during transcript analysis.\n
  • Ingestion points: Call transcripts and summaries are retrieved from the https://stablephone.dev/api/call endpoint (SKILL.md).\n
  • Boundary markers: No specific delimiters or instructions are provided to the agent to distinguish between the transcript content and its own instructions.\n
  • Capability inventory: The agent can execute shell commands via the agentcash CLI for API interactions.\n
  • Sanitization: The skill does not implement sanitization or validation of the call transcripts before they are processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 4, 2026, 06:30 AM