The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill provides instructions for installing the official Anthropic SDK via standard package managers (pip install anthropic and npm install @anthropic-ai/sdk). It also links to official GitHub repositories under the 'anthropics' organization and documentation on 'anthropic.com'. All external references are to well-known, trusted sources.
[CREDENTIALS_UNSAFE]: The skill demonstrates safe credential handling by advising the use of environment variables (ANTHROPIC_API_KEY) and specifically includes examples showing how to retrieve them via 'os.environ.get' and 'process.env'. It correctly identifies hardcoding API keys as a common pitfall.
[COMMAND_EXECUTION]: The shell commands provided in the documentation are restricted to standard package installation and environment variable configuration ('export'). There are no instances of arbitrary or dangerous command execution.
[DATA_EXFILTRATION]: Network operations are exclusively directed at official Anthropic API endpoints. There is no evidence of sensitive data being sent to unauthorized third-party domains.
[PROMPT_INJECTION]: The skill contains standard operational instructions and does not include any patterns typical of prompt injection, such as instructions to override system prompts or bypass safety filters.
[SAFE]: The skill demonstrates secure coding patterns for common tasks including vision model integration, function calling, and web framework integration (FastAPI/Express).

anthropic