The Agent Skills Directory

[SAFE]: The skill is strictly focused on domain modeling and architectural documentation. It lacks any functionality for network communication, external downloads, or execution of system commands.
[SAFE]: No obfuscation, privilege escalation, or persistence mechanisms were found within the skill's instructions or configuration files.
[INDIRECT_PROMPT_INJECTION]: The skill identifies a potential attack surface by ingesting external project documentation.
Ingestion points: Files including CONTEXT.md, CONTEXT-MAP.md, and architectural decision records in docs/adr/ are read into the agent's context.
Boundary markers: The skill does not specify explicit delimiters or instruction-bypass warnings when reading these files.
Capability inventory: The skill's actions are restricted to reading files and creating or updating markdown documentation. It lacks the ability to execute code or perform network operations.
Sanitization: Content from project files is processed as natural language without programmatic sanitization or validation beyond formatting rules.

domain-modeling