typescript-expert
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides instructions and references for TypeScript development. It correctly recommends using the project's locally installed TypeScript version and package manager (npm, pnpm, yarn, bun) to execute commands, adhering to the principle of using the project's own toolchain as the source of truth.\n- [SAFE]: All external links point to official and trusted resources, including the official TypeScript handbook, TSConfig reference, and Microsoft's GitHub repository for TypeScript. These are documented neutrally and serve legitimate educational purposes.\n- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external, potentially untrusted data such as project source code, package.json, and tsconfig.json files. While the skill has the capability to modify files and execute bash commands based on this data, the instructions emphasize minimal changes and verification of fixes. This represents a standard attack surface for developer agents rather than a specific vulnerability in the skill's logic.\n
- Ingestion points: Reads project configuration files (package.json, tsconfig.json, lockfiles) and source code via Read, Glob, and Grep tools in commands/typescript-fix.md.\n
- Boundary markers: None explicitly defined in the prompts to delimit untrusted data from instructions, but the skill maintains a technical 'expert' persona.\n
- Capability inventory: The skill utilizes Edit, Write, and Bash tools to modify project files and run the compiler.\n
- Sanitization: No specific sanitization logic is present; the agent relies on its internal reasoning to interpret the code and configuration.
Audit Metadata