code-execution
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute a local shell script (
setup.sh) to install the required runtime. This script is located within the author's plugin directory (~/.claude/plugins/marketplaces/mhattingpete-claude-skills/). - [DYNAMIC_EXECUTION]: The skill's primary purpose is to enable the agent to generate and execute Python code locally. It uses a specialized library (
execution_runtime) to perform complex codebase transformations and metadata-only analysis with high efficiency. - [INDIRECT_PROMPT_INJECTION]:
- Ingestion points: The skill processes local source files through various file-read and code-analysis functions (e.g.,
fs.copy_lines,read_file). - Boundary markers: None. Untrusted code content is read directly from the filesystem into variables or the agent's context.
- Capability inventory: The skill possesses powerful write capabilities, including filesystem modification (
fs.paste_code,write_file,transform.rename_identifier) and Git operations (git_add,git_commit,git_push). - Sanitization: No sanitization or validation of the content read from files is performed before it is processed or used in subsequent operations.
Audit Metadata