svix-integration-plan
Pass
Audited by Gen Agent Trust Hub on Jun 15, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Fetches documentation and technical specifications from the official Svix documentation site (
docs.svix.com) and public GitHub repositories. These downloads are used to inform the integration plan with up-to-date SDK syntax and feature overviews. - [PROMPT_INJECTION]: The skill is designed to analyze arbitrary local source code to pre-fill integration details, which introduces a surface for indirect prompt injection.
- Ingestion points: Local repository files accessed via
Read(./**)andGrep(./**)as described inSKILL.mdandreferences/triage.md. - Boundary markers: None present; the instructions do not specify delimiters to separate analyzed code from the agent's instructions.
- Capability inventory:
WebFetch,Read,Grep,Glob,AskUserQuestion. - Sanitization: No evidence of sanitization or filtering of the content read from the repository. Note: This risk is inherent to the skill's purpose of codebase analysis and is mitigated by the fact that the output is a document for human review.
- [SAFE]: The skill is authored by Svix and its operations are restricted to relevant vendor domains and official GitHub repositories. No evidence of obfuscation, hardcoded credentials, or persistence mechanisms was found.
Audit Metadata