The Agent Skills Directory

[DATA_EXFILTRATION] (SAFE): The skill does not contain hardcoded credentials or network operations. Documentation in SKILL.md correctly advises the use of environment variables (process.env.ELASTICSEARCH_URL, process.env.TYPESENSE_API_KEY) for backend authentication.
[REMOTE_CODE_EXECUTION] (SAFE): No remote script downloads or execution patterns (e.g., curl | bash) were found. The code uses a static relative import for its internal search module.
[COMMAND_EXECUTION] (SAFE): The CLI handler in index.ts strictly parses commands and does not invoke shell processes or arbitrary system commands.
[PROMPT_INJECTION] (SAFE): The skill instructions are focused on configuration and do not contain directives to bypass AI safety filters or override agent behavior.
[DYNAMIC_EXECUTION] (LOW): The skill uses await import('../../../search/index') in index.ts to load the search module. While dynamic imports can be misused, this instance uses a hardcoded relative path to a local dependency, which is common in modular agent architectures.

search-config