external-tools

Delegate implementation and adversarial review to external AI CLI tools.

Updated Feb 19, 2026
One-click install
npx skills add https://github.com/Prekzursil/quality-zero-platform --skill external-tools-prekzursil
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: external-tools
Source: https://github.com/Prekzursil/quality-zero-platform/tree/main/.claude/plugins/metaswarm/skills/external-tools
Command: npx skills add https://github.com/Prekzursil/quality-zero-platform --skill external-tools-prekzursil

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Delegates implementation and review tasks to external AI CLI tools to reduce cost, eliminate single-model blind spots, and maintain strict safety and validation by having the orchestrator independently verify results.

Core Features & Use Cases

  • One-job-per-invocation: external adapters either implement or review, and never self-certify results.
  • Cross-model adversarial review: writer and reviewer are different models (Codex, Gemini, Claude) to catch blind spots.
  • Safe sandboxing & minimal permissions: isolated git worktrees, env -i minimal environment, and scoped sandboxes for executions.
  • Budgeting & escalation: per-task and per-session circuit breakers with availability-aware escalation chains and health checks.
  • Use Case: offload small feature implementations or remediation branches to cheaper external models, then run automated CI gates and cross-model reviews before merging.

Quick Start

Use the external-tools skill to run an external model implementer on an isolated git worktree with a local spec, perform cross-model adversarial review, and return a reviewed branch if all gates pass.

Frequently Asked Questions about external-tools

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate cross-model code review for feature branches?

Cross-model code review is automated by delegating implementation and adversarial review to different external AI CLI tools, ensuring the writer and reviewer are separate models to catch blind spots before merging.

What is the best way to isolate AI code generation tasks in a sandbox?

AI code generation tasks are isolated using git worktrees and a minimal, scoped sandbox environment via env -i. This restricts permissions and separates the execution context from the main repository branch.

Can I use Codex and Gemini to implement features and have Claude verify them?

Yes, Codex and Gemini adapters can be used to implement features, with Claude acting as a fallback or independent reviewer. This cross-model validation ensures the orchestrator independently verifies results rather than self-certifying.

How do I enforce per-task token or API budgets when running external AI tools?

Per-task and per-session budgets are enforced using built-in circuit breakers and availability-aware escalation chains. The system performs health checks and emits structured JSON facts to track usage against these limits.

Does cross-model external AI delegation work for CI pipeline remediation branches?

Yes, external AI delegation supports repository change tasks like implementing features, bug fixes, and remediation branches. It runs automated CI gates and cross-model reviews before returning a reviewed branch for merging.