Back to list
stephenwinters81

forge

by stephenwinters81

0🍴 0📅 Jan 13, 2026

SKILL.md


name: forge description: Lean code implementation pair. Claude (Smith) writes code, Gemini (Temper) reviews for bugs, security, and edge cases. ~2x token cost for high-quality code output.

The Forge: Code Implementation Pair

When to Activate

Use this skill when the user requests:

  • Code implementation tasks
  • Bug fixes requiring quality review
  • Feature additions with security considerations
  • Any coding task where review overhead is acceptable
  • User explicitly mentions "forge", "pair programming", or "implement with review"

Architecture

Pair: Smith + Temper

The Smith (Claude)

  • Role: Implementer
  • Writes code based on user requirements
  • Follows existing patterns in codebase
  • Produces complete, working implementation

The Temper (Gemini 3 Pro)

  • Role: Adversarial Reviewer
  • Reviews Smith's code for bugs
  • Checks for security vulnerabilities
  • Identifies edge cases and missing error handling
  • Returns PASS or FAIL with specific issues

Execution Flow

[User Request]
    ↓
[Smith] → Writes implementation
    ↓
[Temper] → Reviews code
    ↓
    ├── PASS → Return code to user
    └── FAIL → Smith fixes issues → Temper re-reviews

Invocation

python3 ~/.claude/skills/forge/forge.py "Your implementation task"

Options

python3 ~/.claude/skills/forge/forge.py "Your task" --max-iterations 3 --verbose

Example Usage

User: "Add input validation to the login form"

python3 ~/.claude/skills/forge/forge.py "Add input validation to the login form in web/components/auth/LoginForm.tsx - validate email format and password length"

Why This Works

  1. Cross-Model Review: Gemini catches Claude's blind spots
  2. Lean Overhead: Only 2 agents vs 3+ in other formations
  3. Focused Scope: No planning phase - just implement and review
  4. Iterative Fix: Failed reviews trigger targeted fixes

Token Efficiency

  • ~2x base cost (1 implementation + 1 review)
  • Best for: Medium complexity, single-file or few-file changes
  • Avoid for: Large architectural changes (use Foundry instead)

Requirements

  • Claude Code CLI installed and authenticated (claude command available)
  • Gemini CLI installed (gemini command available)
  • Python package: rich (for formatted output)

Score

Total Score

50/100

Based on repository quality metrics

SKILL.md

SKILL.mdファイルが含まれている

+20
LICENSE

ライセンスが設定されている

0/10
説明文

100文字以上の説明がある

0/10
人気

GitHub Stars 100以上

0/15
最近の活動

3ヶ月以内に更新がある

0/10
フォーク

10回以上フォークされている

0/5
Issue管理

オープンIssueが50未満

+5
言語

プログラミング言語が設定されている

+5
タグ

1つ以上のタグが設定されている

0/5

Reviews

💬

Reviews coming soon