The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill creates an attack surface by ingesting and processing data from external sources during its troubleshooting and functional check workflows.
Ingestion points: The agent processes output from the auth_me and list_inboxes tools, as well as the content of a remote machine-readable guide located at https://api.agentmailer.ai/auth.md (referenced in troubleshooting.md).
Boundary markers: Absent; there are no specific instructions for the agent to use delimiters or ignore embedded instructions when processing these external data sources.
Capability inventory: The skill utilizes shell command execution via the claude mcp add and codex mcp add commands for environment configuration.
Sanitization: No explicit sanitization, validation, or filtering mechanisms are defined for the data retrieved from the remote MCP server or the auth guide.

agentmailer-mcp