The Agent Skills Directory

[COMMAND_EXECUTION]: The skill requires the agent to execute shell commands for building, compiling, and validating code implementation.
Evidence: SKILL.md contains instructions such as "代码可编译/可构建/可运行" (Code is compilable/buildable/runnable) and "运行相关验证" (Run related validations). It also references executing generation and export scripts defined in project configuration.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because its core logic relies on processing and following instructions from external project documentation.
Ingestion points: The agent reads data from change.md, spec.md, test.md, review.md, implement-layers.md, and project templates in the eo-doc/ directory.
Boundary markers: There are no explicit markers or instructions to treat the content of these files as untrusted data or to ignore embedded agent instructions.
Capability inventory: The agent has the ability to write to the file system (creating implementation files and updating change.md) and execute system commands (compilation and testing).
Sanitization: The skill lacks mechanisms to sanitize or validate the content of the documentation files before implementation or execution steps.

eo-implement