The Agent Skills Directory

[SAFE]: The skill uses designated Apollo MCP tools to perform lead generation tasks as described. No malicious code, obfuscation, or unauthorized exfiltration patterns were identified.
[PROMPT_INJECTION]: The skill parses user input from the $ARGUMENTS variable to define search filters. This is a standard functional requirement for this skill but represents a surface where adversarial input could attempt to influence the agent's behavior.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from user input and external API results which could potentially contain instructions.
Ingestion points: $ARGUMENTS in SKILL.md and data returned from mcp__claude_ai_Apollo_MCP tools.
Boundary markers: None present.
Capability inventory: Searching, enriching, and saving contacts via Apollo MCP tools.
Sanitization: No explicit sanitization of data retrieved from the Apollo API is implemented.

prospect