The Agent Skills Directory

[SAFE]: The skill serves as a static analysis and code review guide for HarmonyOS development. It instructs the agent to use standard tools like rg (ripgrep) to identify potential issues in project files. All identified URLs point to official Huawei documentation, which are recognized as trusted/well-known services.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted source code from HarmonyOS repositories, which represents an attack surface for indirect prompt injection.
Ingestion points: Reads .ets, .ts, and .json5 files from the working tree as described in SKILL.md.
Boundary markers: The instructions include strong directives to verify all findings against reachable code and official documentation, and to distinguish production code from examples or test fixtures to mitigate accidental obedience to embedded instructions (SKILL.md).
Capability inventory: Uses the rg (ripgrep) utility for text searching within the repository.
Sanitization: The skill relies on strict evidence-based reporting requirements, requiring precise file and line references and verification against platform contracts before any finding is reported.

harmonyos-review