The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill provides instructions for the agent to process external CSV files using the LeadMagic CLI, which creates a vulnerability surface for instructions embedded in data to influence the agent.
Ingestion points: The skill frequently references ingesting external data via files like contacts.csv and external-contacts.csv using the lm enrich and lm validate commands, as seen in SKILL.md and references/cli-workflow-patterns.md.
Boundary markers: The instructions do not define any specific delimiters or "ignore instructions" warnings to protect the agent from processing malicious content hidden within the CSV rows.
Capability inventory: The skill allows for significant capabilities including email finding, contact validation, and pushing data to outbound platforms like Smartlead and Instantly.
Sanitization: The accompanying validation script (scripts/check-output.py) focuses on checking for required markdown sections and keywords in the deliverable but does not perform any sanitization or validation of the actual data content processed by the CLI commands.

leadmagic-cli