skills/cursor/plugins/Poteto Mode/Gen Agent Trust Hub

Poteto Mode

Pass

Audited by Gen Agent Trust Hub on Aug 21, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the Git, GitHub CLI (gh), and Graphite CLI (gt) tools to manage code repositories and pull request workflows. These operations are essential for its purpose as a development assistant and are performed using standard process spawning methods.
  • [EXTERNAL_DOWNLOADS]: The scripts/bootstrap.ts file includes a routine to automatically install necessary Node.js dependencies using bun install --frozen-lockfile. This is a common and safe dependency management practice for developer tools provided by trusted vendors.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources, including GitHub PR threads and automated bot comments. It mitigates this attack surface by explicitly instructing the agent to treat comment text as untrusted data and providing a skeptical triage rubric in references/bugbot-triage.md to prevent external input from overriding agent behavior.
  • [DYNAMIC_EXECUTION]: The skill uses a self-restarting pattern in its bootstrap script to ensure that the environment is correctly configured after dependency installation. It also employs a structured 'brief' system to spawn subagents for parallel work, which includes passing explicit goals and constraints to maintain operational safety.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 21, 2026, 07:24 PM
Security Audit — agent-trust-hub — Poteto Mode