The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides numerous Bash and Python scripts that perform local file system operations, including directory creation (mkdir -p), writing logs to system paths (/var/log/), and managing local file assets.- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface where untrusted data is processed. Ingestion points: Data is ingested via script arguments ($INPUT_TEXT in conditional_workflow.sh) and local file content (cat $file in data_processing.sh). Boundary markers: No delimiters or XML tags are used to isolate untrusted data from the instructions. Capability inventory: The skill uses the infsh CLI to send this data to external AI models and the Bash tool for orchestration. Sanitization: No validation or escaping is performed on the input data before interpolation into prompts.- [EXTERNAL_DOWNLOADS]: The skill references an external installation script for the infsh CLI hosted on GitHub and suggests installing additional functional skills using npx (e.g., npx skills add inference-sh/skills@ai-content-pipeline), which involves downloading and executing remote packages.- [DATA_EXFILTRATION]: The monitoring and alerting examples demonstrate the use of curl to send execution logs, error messages, and command outputs to external webhooks (e.g., your-webhook.com).

ai-automation-workflows