The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted data from source code and OpenAPI specifications to generate documentation and drive sub-agent tasks. \n- Ingestion points: OpenAPI YAML files (e.g., openapi/*.yaml) and source code directories (e.g., apps/*/src/). \n- Boundary markers: The skill lacks explicit instructions to treat code or specifications as data only or to ignore embedded natural language instructions. \n- Capability inventory: The skill executes local Git commands, performs file system write operations to documentation directories, and spawns sub-agents for parallel processing. \n- Sanitization: No validation or sanitization of content extracted from source files is specified before it is used in documentation or passed to sub-agents.\n- [COMMAND_EXECUTION]: The skill executes local shell commands to facilitate version tracking and change detection. Specifically, it uses git rev-parse, git log, and git diff. These commands are standard for the skill's purpose and operate on local paths defined in the service configuration table.

sync-service-docs