The Agent Skills Directory

[SAFE]: The skill provides a standard interview and modeling methodology for integrating external systems. It uses vendor-specific storage paths in the .trogonai directory for maintaining project documentation, which is consistent with its stated purpose.
[NO_CODE]: There are no executable scripts, binary files, or network-enabled code blocks within the skill. The logic is purely instructional for the AI agent's behavior during a manual modeling session.
[INDIRECT_PROMPT_INJECTION]: The skill requires the agent to ingest and analyze sample payload data from external systems (such as webhooks or API responses) provided by the user. This creates a surface where instructions embedded within these technical samples could potentially influence the agent's behavior.
Ingestion points: Sample payloads and technical field definitions are collected from the user via the AskUserQuestion tool during the interview and workflow phases.
Boundary markers: The instructions do not provide explicit delimiters or "ignore" directives to isolate external payload samples from the agent's processing logic.
Capability inventory: The agent is authorized to use the Write tool to modify or create files in the local filesystem.
Sanitization: No specific sanitization or validation steps are defined for the technical data extracted from external systems before it is processed or written to disk.

eventmodeling-translating-external-events