agentic-design-patterns-chinese
Warn
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides instructions to clone a repository from an untrusted GitHub account at
https://github.com/xindoo/agentic-design-patterns.git. - [REMOTE_CODE_EXECUTION]: Local setup instructions require running
bundle installandnpm install -g gitbook-cli, which download and execute code from public registries. - [COMMAND_EXECUTION]: The documentation includes multiple commands for shell execution, including project setup (
git,npm,bundle), server management (jekyll,gitbook), and content processing usinggrepandsed. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external markdown files in the
chapters/directory, creating an attack surface for instructions embedded in data. 1. Ingestion points: Markdown files in thechapters/directory (SKILL.md). 2. Boundary markers: No delimiters or warnings are specified for processing these files. 3. Capability inventory: Shell command execution (git,npm,bundle,jekyll,gitbook,grep,sed) and file system access. 4. Sanitization: No sanitization of input data is described before extraction or processing.
Audit Metadata