The Agent Skills Directory

[COMMAND_EXECUTION]: The SKILL.md file instructs the agent to execute a shell command to run a local Node.js script (sync.js) and perform git diff on the results. This is the intended primary function of the skill to automate file synchronization.
[INDIRECT_PROMPT_INJECTION]: The skill modifies configuration files (AGENTS.md, CLAUDE.md) that are used by agents to define their behavior and standards. This creates a surface where instructions from the skill's source files are persisted into the agent's operating context.
Ingestion points: The script sync.js reads content from references/standards.md located within the skill directory.
Boundary markers: The script injects content under specific Markdown headers (e.g., ## 交流语言, ## 写作要求) but does not include explicit warnings to ignore embedded instructions within those sections.
Capability inventory: The skill uses sync.js to perform file system reads (fs.readFileSync) and writes (fs.writeFileSync). It also performs shell execution for the sync script and git diff as described in SKILL.md.
Sanitization: The content from the source file is injected verbatim into the target files without escaping or sanitization of potential instructions.

sync-writing-standards