The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it ingests untrusted data that is then used to generate content for automated email campaigns via the Instantly API.
Ingestion points: User-provided inputs including client_name, client_description, offers, target_audience, and social_proof are passed via CLI arguments to scripts/instantly_create_campaigns.py.
Boundary markers: The prompt template in generate_campaigns_with_claude interpolates these variables directly into the instruction block without any delimiters (e.g., XML tags or triple quotes) or 'ignore embedded instructions' warnings.
Capability inventory: The skill has the capability to create live email campaigns in Instantly using requests.post to https://api.instantly.ai/api/v2/campaigns.
Sanitization: There is no sanitization or validation of the input strings to detect or neutralize malicious instruction injection prior to prompt construction.

instantly-campaigns