iteration-manager
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill invokes local command-line tools for development tasks, including pytest for test execution, ruff and flake8 for linting, and mypy for type checking. These are standard tools for the skill's purpose of code quality assurance.
- [DYNAMIC_EXECUTION]: The skill implements an automated cycle where it generates code modifications to fix bugs or improve coverage and applies these changes to the project files. This behavior is the central function of the iteration manager.
- [INDIRECT_PROMPT_INJECTION]: The skill processes external project code and test output strings which are untrusted. This represents an attack surface for indirect prompt injection.
- Ingestion points: Project source code files and the standard output/error streams from test and linting tools.
- Boundary markers: No explicit delimiters or instructions are used to separate untrusted tool output from the agent's internal reasoning.
- Capability inventory: The agent has the ability to execute shell commands and write to files on the local system.
- Sanitization: No evidence of filtering or sanitizing the content read from files or tool outputs before it is processed by the agent.
Audit Metadata