The Agent Skills Directory

[SAFE]: The skill interacts with the official LeadMagic API endpoints (api.leadmagic.io) to fetch advertisement data. This is consistent with the skill's stated purpose of providing ads intelligence.
[SAFE]: Authentication is managed using the LEADMAGIC_API_KEY environment variable in the provided examples, which is a standard and safe practice for handling secrets.
[INDIRECT_PROMPT_INJECTION]: The skill ingests advertisement copy from external API sources, which represents a potential surface for indirect prompt injection. However, the instructions provide specific rules to only report creatives the API returned and not to invent or paraphrase, which mitigates the risk of the agent obeying malicious instructions embedded in ad data.
Ingestion points: API responses from v1/ads endpoints in SKILL.md.
Boundary markers: Absent; no specific delimiters are defined for the fetched content.
Capability inventory: Network requests (via curl or MCP tools) to api.leadmagic.io.
Sanitization: Not explicitly specified for the fetched advertisement text.

ads-intelligence