cangjie-dev
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill acts as a Cangjie developer assistant and is activated by source files and project configuration, which creates a data ingestion surface.
- Ingestion points: Source files (.cj) and project configurations (cjpm.toml) are processed to provide context for code generation and assistant behavior (README.md, SKILL.md).
- Boundary markers: The skill instructions do not define delimiters or specific warnings for the agent to ignore natural language instructions embedded in code comments or metadata.
- Capability inventory: The skill is capable of code generation, completion, and providing build and test command guidance for the local environment (SKILL.md).
- Sanitization: There is no explicit logic to filter or escape potentially malicious instructions that could be found in the processed project data.
Audit Metadata