The Agent Skills Directory

[PROMPT_INJECTION] (LOW): Indirect Prompt Injection surface detected through data ingestion and file modification.
Ingestion points: The skill is designed to read user-provided .ets source files and contents from references/error-log.md.
Boundary markers: Absent. The instructions do not specify delimiters or safety warnings to distinguish between instructions and data when processing external file content.
Capability inventory: The agent uses Grep for searching files and Edit for writing/appending to references/error-log.md.
Sanitization: Absent. User-provided error messages and code snippets are recorded into the long-term error-log.md without validation or escaping, which could lead to persistent instruction injection in future sessions.
[COMMAND_EXECUTION] (SAFE): The skill references running compilation commands as part of the development workflow. This behavior is consistent with the primary purpose of a development tool and is gated by a reference guide.

hmos-development