The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes a validation script (scripts/check-output.py) designed to be run locally by the user or agent to ensure deliverables meet quality standards. The script performs basic regex checks for required sections and keywords and alerts the user if hardcoded API keys are detected in the output files.
[DATA_EXPOSURE]: The instructions explicitly forbid hardcoding API keys and sensitive credentials in configuration files, recommending the use of environment variables or client secret stores. This aligns with industry best practices for secret management.
[PRIVILEGE_ESCALATION]: The skill implements a 'Least Privilege' approach by distinguishing between read-only enrichment tools and 'side-effect' tools (CRM writes). It mandates explicit human confirmation before the agent is permitted to perform any write operations or enroll contacts in external sequences.

leadmagic-mcp