The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses a Bash wrapper (scripts/search) to execute a Node.js script (scripts/search.mjs). This is a standard pattern for CLI-based skills and does not involve arbitrary command injection or shell escapes.- [CREDENTIALS_UNSAFE]: The skill manages a Valyu API key. It correctly avoids hardcoding secrets and instead uses a local configuration file (~/.valyu/config.json) or environment variables (VALYU_API_KEY). This follows security best practices for credential management in development tools.- [EXTERNAL_DOWNLOADS]: The skill communicates with api.valyu.ai to perform semantic searches. This is the primary and intended function of the skill. No other external downloads or remote script executions were identified.- [DATA_EXFILTRATION]: Network activity is restricted to the Valyu API endpoint for search queries. There is no evidence of sensitive local files (like SSH keys or AWS credentials) being read or transmitted to external servers.- [PROMPT_INJECTION]: The skill documentation uses standard instructional language. It does not contain patterns intended to bypass AI safety filters or override system prompts.

drugbank-search