The Agent Skills Directory

[SAFE]: The skill is designed to interact with the official LeadMagic API endpoints at api.leadmagic.io and mcp.leadmagic.io for legitimate business data enrichment tasks.
[SAFE]: Authentication is correctly handled using the $LEADMAGIC_API_KEY environment variable placeholder, ensuring that no sensitive credentials are hardcoded within the skill instructions.
[SAFE]: No obfuscation, persistence mechanisms, or unauthorized privilege escalation attempts were found. The use of shell commands is limited to standard curl requests for API communication.
[INDIRECT_PROMPT_INJECTION]: While the skill ingests external data such as profile URLs and job titles, it functions as a structured API client. The risk is minimized as data is sent to a specific vendor endpoint rather than being interpreted as instructions by the agent.
Ingestion points: User-provided inputs like profile_url, job_title, and company_domain in SKILL.md.
Boundary markers: Not explicitly defined in this instructional file.
Capability inventory: Network operations via curl to api.leadmagic.io.
Sanitization: None specified in the instructions; sanitization is expected to be handled by the LeadMagic API server-side.

people-enrichment