The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes several shell commands for development workflows, including terraform-docs for generating documentation, uv run for executing local validation scripts, rg (ripgrep) for searching variables, and terraform fmt for code formatting. These are standard operations for managing Infrastructure-as-Code projects.
[EXTERNAL_DOWNLOADS]: The skill utilizes the gemini CLI tool to process file content via Google's Gemini models. This involves transmitting local file metadata and content (e.g., from variables.tf) to a well-known service for analysis, which is the intended purpose of the skill.
[INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection as it processes untrusted data from repository files.
Ingestion points: Reads content from variables.tf, docs/llms.md, kube.tf.example, and README.md to identify gaps and update documentation (Steps 1, 2, 5, and 7).
Boundary markers: Absent; the instructions do not use specific delimiters or instructions to ignore instructions embedded within the source files.
Capability inventory: The skill has the capability to execute shell commands (uv run, gemini, terraform-docs, rg) and perform file system writes to update repository documentation.
Sanitization: No explicit sanitization or escaping is performed on the ingested file content before it is processed by the CLI tools. However, in the context of a documentation synchronization tool for a specific project, this is a standard operating risk.

sync-docs