The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local Python scripts (scripts/token_store.py and scripts/adjust_report.py) to manage API tokens and process reporting data.
[CREDENTIALS_UNSAFE]: The skill requests and stores sensitive Adjust API and App tokens. These are saved to a local configuration file (~/.vk-cowork/adjust_credentials.json). The script explicitly sets file permissions to 0600 (read/write only for the current user) to mitigate unauthorized local access.
[EXTERNAL_DOWNLOADS]: The skill retrieves data from automate.adjust.com. This is the official and well-known domain for Adjust's automated reporting services.
[PROMPT_INJECTION]: The skill processes campaign names and other metadata from external API responses. This represents a potential surface for indirect prompt injection, which is a common risk when agents process external data, though no specific malicious patterns were found in the ingestion logic.

adjust-report