The Agent Skills Directory

[SAFE]: The skill interacts with the Abel API at 'cap.abel.ai' to retrieve causal graph data. This is the primary function of the agent and is handled via 'scripts/cap_probe.py' using standard Python libraries.- [SAFE]: Authorization keys for the Abel API and GitHub are managed securely using environment variables or local '.env' files. The skill includes a setup guide for an OAuth-based key retrieval process.- [SAFE]: The update mechanism in 'scripts/check_skill_update.py' checks for new versions of the skill from the author's official GitHub repository. It uses 'subprocess' to interact with established developer tools like the GitHub CLI and the 'skills' NPM package.- [SAFE]: A dedicated rendering guard script ('scripts/render_guard.py') is used to ensure that internal data identifiers and raw numeric predictions are filtered out of the final report, enhancing the privacy and readability of the output.- [SAFE]: The skill incorporates a robust web-grounding process with specific search strategies and source hierarchy rules. These instructions ensure that information from the web is cross-referenced with causal graph data before being presented to the user.

causal-abel