The Agent Skills Directory

[COMMAND_EXECUTION]: The skill makes extensive use of the oma CLI for workflow management, including commands such as oma state emit to record decisions, oma state verify for checkpointing, oma diagram resolve for architectural visualization, and oma agent spawn for cross-agent consultation.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the user's codebase and documentation to perform architecture reviews. The results of this analysis are used to construct arguments for shell commands, which presents a surface for potential injection if malicious content in the processed files influences the generated recommendation.
Ingestion points: The skill reads source files and documentation within the CODEBASE scope.
Boundary markers: No explicit delimiter or "ignore" instructions are defined for the reading of external codebase files.
Capability inventory: The skill has the ability to execute shell commands via the oma CLI and perform file system writes to the .agents/results/architecture/ directory.
Sanitization: There are no explicit instructions to escape or validate data derived from the codebase before it is interpolated into shell command templates like oma state emit.
[EXTERNAL_DOWNLOADS]: The skill's oma diagram resolve command is documented to automatically fetch the latest release of the archify diagram engine when it is selected for architectural visualization. This is a standard update mechanism for the vendor's platform tools.

oma-architecture