The Agent Skills Directory

[PERSISTENCE]: The skill implements a persistence mechanism by allowing the user to install a recurring schedule on the host operating system. The install_cron.sh script configures a launchd agent on macOS or a crontab entry on Linux to execute the run_snapshot.py script at specified intervals. This behavior is documented as the core functionality for automated visibility tracking.
[COMMAND_EXECUTION]: The workflow involves the execution of multiple shell scripts (install_cron.sh, uninstall_cron.sh) to manage the system scheduling. Additionally, the Python runner (run_snapshot.py) executes logic for API communication, data processing, and local file generation.
[EXTERNAL_DOWNLOADS]: The skill requires the installation of external Python libraries, specifically requests and tldextract, which are fetched from public package registries. It also invokes remote Apify Actors (apify/google-search-scraper, apify/website-content-crawler) to perform search and crawling tasks.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from AI search engines, which is interpolated into Markdown reports. This represents an attack surface where malicious content on cited websites could attempt to influence the agent's behavior during report analysis. Ingestion points include run_snapshot.py via fetch_dataset_items. Capabilities include network access to Apify API and local file system writes for report storage. The skill uses structured JSON parsing as a boundary, but remains vulnerable to the content within those fields.

apify-ai-search-visibility-tracker