x-ray
Pass
Audited by Gen Agent Trust Hub on Apr 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local bash and python scripts (
enumerate.sh,analyze_git_security.py,generate_svg.py) to analyze codebase statistics, git history, and generate diagrams. It also invokes external development toolchains likeforge(Foundry) andnpx hardhatto gather coverage data. - [EXTERNAL_DOWNLOADS]: Fetches a version file from the author's public GitHub repository (
raw.githubusercontent.com/pashov/skills/main/x-ray/VERSION) to check for updates. - [PROMPT_INJECTION]: Indirect prompt injection surface exists because the skill ingests and processes user-provided Solidity source code. While the skill instructs subagents to perform objective fact extraction, malicious content within source files could theoretically attempt to influence the automated reporting phase.
- Ingestion points: Solidity source files read during Step 2 in
SKILL.md. - Boundary markers: The prompt explicitly instructs subagents to "Do NOT analyze — just extract facts."
- Capability inventory: Shell command execution (bash, python3), file system writing, and subagent invocation.
- Sanitization: No specific input sanitization or filtering of Solidity code content is detailed.
Audit Metadata