The Agent Skills Directory

[PROMPT_INJECTION]: No malicious injection patterns or behavioral overrides were detected. The skill instructions prioritize security boundaries and least-privilege principles.- [DATA_EXFILTRATION]: No exfiltration patterns were found. The skill explicitly warns against logging or transmitting secrets, and external links point to official documentation and RFC repositories.- [COMMAND_EXECUTION]: The included Python script scripts/validate_tool_manifest.py performs static analysis of JSON manifest files using only standard library modules and does not invoke external commands or network resources.- [CREDENTIALS_UNSAFE]: No hardcoded credentials were found. The skill provides clear guidance on safe credential management, emphasizing the use of environment variables and secret stores instead of embedding secrets in source code or arguments.- [REMOTE_CODE_EXECUTION]: No remote code execution patterns or unverified external downloads were detected.

mcp-server-building