The Agent Skills Directory

[SAFE]: The skill implements standard architectural modeling workflows. All included scripts and commands are restricted to the skill's stated purpose of managing LikeC4 models.\n- [EXTERNAL_DOWNLOADS]: The skill utilizes the official likec4 CLI via npx for model validation and diagram rendering. This is a well-known tool for architecture-as-code.\n- [INDIRECT_PROMPT_INJECTION]: The skill translates natural language descriptions into LikeC4 DSL. While it ingests external data (user queries), the scope is limited to generating code for the LikeC4 toolchain, and the output is validated by a dedicated CLI. Evidence chain: Ingestion occurs via user prompts; Boundary markers are absent; Capability inventory includes file-system writing and subprocess execution (npx likec4); Sanitization is handled by CLI validation.

likec4-architecture