The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute a local bash script (setup-architect.sh) found within the project's .agents directory to initialize the environment for ADR documentation. This is an expected behavior for the vendor's architectural tool suite.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied Product Requirements Documents (PRDs) to guide architectural discovery and documentation, creating a potential surface for indirect injection.
Ingestion points: Product requirements provided by the user as arguments to the skill.
Boundary markers: None specified in the instructions to distinguish untrusted requirement text from the agent's instructions.
Capability inventory: File writing to the repository (.adlc/drafts/adr/) and execution of a local shell script.
Sanitization: No input validation or sanitization is performed on the ingested PRD content.

architect-specify