The Agent Skills Directory

[SAFE]: The skill provides legitimate technical documentation and implementation patterns for the Apollo.io API.
[COMMAND_EXECUTION]: Uses standard shell utilities (curl, jq, wc, xxd) for API health checks and formatting diagnostic output.
[EXTERNAL_DOWNLOADS]: Interacts with official Apollo.io domains (api.apollo.io, status.apollo.io) for connectivity verification and status monitoring.
[PROMPT_INJECTION]: The skill identifies an indirect prompt injection surface. Untrusted data enters the context via API response bodies (ingestion points: err.response.data in SKILL.md and references/implementation-guide.md). Boundary markers and sanitization logic are absent in the provided code snippets. The agent environment possesses capabilities including Bash, curl, Grep, and Read that could theoretically be influenced by external content if the trusted API source were compromised.
[CREDENTIALS_UNSAFE]: Instructions include debugging steps that pass API keys via query parameters in cURL commands or echo them to the terminal to check length. These practices facilitate troubleshooting but may expose sensitive credentials in shell logs, proxies, or history if not carefully managed.

apollo-common-errors