code-review-architecture
Warn
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: MEDIUMDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [DYNAMIC_EXECUTION]: The skill generates a Python script for cycle detection at /tmp/cycle_detect.py using a hardcoded template and executes it using python3 to process the project's dependency graph.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from the project's source files and documentation to build its analysis. Ingestion points: Project source code, Markdown documentation (SKILL.md, agents/*.md), and hook configurations (hooks.json). Boundary markers: None are present to delimit untrusted data or instruct the agent to ignore instructions embedded within these files. Capability inventory: Execution of python3, node, uv run, and artifact registration via sam_cli. Sanitization: The skill does not perform sanitization or validation of the content extracted from project files before processing it for graph generation or identifying cross-language invocations.
- [COMMAND_EXECUTION]: The skill executes multiple system commands and CLI tools, including git, grep, glob, ccc, graphify, node, and uv, to analyze the codebase structure and dependencies based on patterns found in the project's files.
Audit Metadata