The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses basic shell commands such as test -n and export to verify and manage the EXTRUCT_API_TOKEN. This is a standard and safe procedure for ensuring required credentials are set before execution.
[EXTERNAL_DOWNLOADS]: It retrieves live documentation from the vendor's official website (https://www.extruct.ai/docs). This is a safe operation used to synchronize local API references with current documentation.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it processes data from API responses and local context files. However, this is inherent to its primary function of data enrichment and no malicious instructions were detected.
Ingestion points: Data retrieved from GET /tables/{table_id}/data and local project context files (claude-code-gtm/context/{company}_context.md).
Boundary markers: None explicitly defined in the workflow instructions.
Capability inventory: API calls to create and run table columns; file writing to specific CSV paths for campaign data.
Sanitization: No explicit sanitization or filtering of external data is noted.

people-search