expo-cicd-workflows
Pass
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- External Resource Retrieval: The skill is designed to fetch the latest EAS workflow schema and syntax documentation from official Expo servers and their public GitHub repository. These downloads are performed to ensure validation and generation processes use current specifications.
- Local Script Execution: The skill includes utility scripts (
fetch.jsandvalidate.js) that are executed using Node.js to automate data retrieval and YAML validation. These scripts rely on well-known libraries likeajvandjs-yamlfor processing data. - Indirect Prompt Injection Surface: The skill processes information from remote URLs and local configuration files, which could serve as an ingestion point for untrusted content.
- Ingestion points: Remote JSON schema and documentation URLs (defined in
SKILL.md) and local workflow YAML files (processed byvalidate.js). - Boundary markers: No specific delimiters or instructions are provided to the agent to ignore potentially malicious embedded instructions in the ingested data.
- Capability inventory: The skill possesses network access capabilities (via
fetch.js), file system read/write permissions for caching and validation, and the ability to execute local Node.js processes. - Sanitization: While the skill uses structured parsers for JSON and YAML to ensure format integrity, it does not implement specific sanitization or filtering for natural language instructions embedded within that data.
Audit Metadata