card-onboard

Pass

Audited by Gen Agent Trust Hub on Jul 16, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the Bash tool to run a local diagnostic script, scripts/probe.sh, which identifies the state of the system daemon, connection transport, and firmware.\n- [COMMAND_EXECUTION]: The scripts/probe.sh script performs several local command executions, including process checks (pgrep), network probes to a local daemon (curl to 127.0.0.1), and execution of a firmware installation script (install_firmware.py).\n- [PROMPT_INJECTION]: The skill presents an indirect prompt injection attack surface (Category 8) by ingesting untrusted user input and script outputs.\n
  • Ingestion points: User-provided Wi-Fi SSID and password in SKILL.md, and JSON data returned by scripts/probe.sh.\n
  • Boundary markers: Absent; the skill does not use delimiters to wrap untrusted input.\n
  • Capability inventory: All subprocess calls, including pgrep, curl, and python3, are executed within scripts/probe.sh, while the Bash tool is used in SKILL.md.\n
  • Sanitization: Absent; user-provided credentials are interpolated into the /card-wifi-setup command in SKILL.md without documented escaping or validation.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 16, 2026, 10:34 AM
Security Audit — agent-trust-hub — card-onboard