The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because its primary function involves the agent reading and interpreting untrusted data from local codebases.
Ingestion points: The analyze_codebase method in the QualityOrchestrator class and the scan method in ProactiveQualityScanner process files from user-defined paths (e.g., src/).
Boundary markers: The skill does not implement explicit boundary markers or "ignore instructions" directives when passing codebase content to the agent, which could lead to the agent inadvertently following instructions embedded in code comments or string literals.
Capability inventory: The skill possesses the capability to read files via the Read tool and can suggest or automatically apply code refactors (modifying the filesystem) based on the analysis of these untrusted files.
Sanitization: There is no evidence of content sanitization or instruction filtering for the codebase data before it enters the agent's context.
[EXTERNAL_DOWNLOADS]: The skill integrates with the Context7 MCP service to retrieve live documentation and best practices guidelines.
Documentation Fetching: It uses the mcp__context7__get-library-docs tool to fetch topics like "best-practices", "performance", and "security".
Trusted Sources: The library mappings reference well-known and trusted technology organizations and repositories, including facebook/jest, google/lighthouse, microsoft/TypeScript, and pytest-dev/pytest.
[SAFE]: The skill utilizes vendor-specific packages (e.g., moai-foundation-quality, moai-core-quality) and well-known industry tools for its intended purpose of code quality management.
Vendor Context: All package references and integration patterns (such as those for GitHub Actions and REST APIs) align with the stated purpose and the identity of the author 'modu-ai'.

moai-foundation-quality