specs-kit-code-cleanup
Pass
Audited by Gen Agent Trust Hub on Aug 28, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructions require the agent to search for and execute scripts defined in project configuration files such as
package.json,Maven/Gradleplugins,composerscripts, andpyproject.toml. If a repository contains a malicious configuration file, the agent will execute those scripts in the local environment. - [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from project task files (
TASK-XXX.md) and the source code itself to determine which files to modify and what technical debt to remove. It lacks explicit boundary markers or sanitization for this content. - Ingestion points:
docs/specs/[id]/tasks/TASK-XXX.md, project source files, and configuration files (package.json,pyproject.toml, etc.). - Boundary markers: None identified in the instruction set for handling external data.
- Capability inventory: File reading, file writing (cleanup), and shell command execution (linting, formatting, testing).
- Sanitization: No evidence of input validation or command sanitization before executing project-defined tools.
Audit Metadata