dense-multi-agent-worktree-delivery
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill frequently executes shell commands using git, pnpm, and python to manage code worktrees, run tests, and perform administrative tasks. These operations are core to the skill's function and are performed within a controlled developer environment.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and act upon data from external sources including Linear issues, Slack messages, and GitHub pull request bodies. This creates a vulnerability to indirect prompt injection where malicious instructions could be embedded in those sources. The skill mitigates this risk by using specific, scoped prompts for sub-agents and enforcing a mandatory review loop.
- [EXTERNAL_DOWNLOADS]: The skill uses npm and homebrew to install development dependencies like pnpm and graphite. These downloads are directed at official and trusted software registries and are necessary for the described developer workflow.
Audit Metadata