compose-next
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: Orchestrates repository management using shell commands for Git worktree creation (
git worktree add), state inspection (git rev-parse), and GitHub CLI integration (gh pr merge). - [EXTERNAL_DOWNLOADS]: Synchronizes project dependencies using
bun cianduv sync --frozen. These commands download packages from well-known official registries (NPM and PyPI). - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted repository data (code, README files, AGENTS.md) which could theoretically contain malicious instructions.
- Ingestion points: Reads project-specific instructions and source code from the repository workspace.
- Boundary markers: The skill does not define specific delimiters for separating repository content from its own instructions.
- Capability inventory: Has the ability to write to the file system, execute shell commands, and dispatch subagents for parallel tasks.
- Sanitization: No explicit sanitization or filtering of codebase content is mentioned before it is processed by the agent.
Audit Metadata