serverchan

Fail

Audited by Gen Agent Trust Hub on Jul 28, 2026

Risk Level: CRITICAL
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill communicates with sctapi.ftqq.com and ft07.com, which are the official and documented endpoints for the ServerChan (方糖) notification service. These are recognized as functional requirements for the skill's purpose.
  • [CREDENTIALS_UNSAFE]: The skill handles a SERVERCHAN_SENDKEY credential but follows security best practices by recommending the use of environment variables and explicitly including logic to avoid printing or logging the key to stdout/stderr.
  • [COMMAND_EXECUTION]: The skill uses bash and curl to perform network operations. It uses --data-urlencode and Python-based JSON assembly in scripts/send.sh to ensure user-provided titles and descriptions are properly escaped, mitigating potential injection risks into the HTTP request body.
  • [DATA_EXFILTRATION]: While the skill sends data to an external service, this is its primary stated purpose (pushing notifications). The data sent is limited to what the user or agent provides as the notification content, and it targets official service domains.
Recommendations
  • Contains 2 malicious URL(s) - DO NOT USE
Audit Metadata
Risk Level
CRITICAL
Analyzed
Jul 28, 2026, 02:47 AM
Security Audit — agent-trust-hub — serverchan