The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill designs research prompts based on external inputs such as hypothesis sets and target lists. If these external sources contain malicious instructions, they could be incorporated into the final column_configs and executed by downstream enrichment agents.
Ingestion points: The skill reads hypothesis sets (from market-research output or context files) and target lists as defined in SKILL.md.
Boundary markers: No explicit delimiters or boundary markers are used to separate user-provided hypotheses from the generated prompt templates.
Capability inventory: The skill generates configurations that trigger research_pro (web research), llm, research_reasoning, and linkedin tools in the list-enrichment workflow.
Sanitization: There is no evidence of filtering or sanitization of input data before it is interpolated into the {prompt} fields of the JSON output.

enrichment-design