tammai

Rubric Judge — Development agent for Claude Code

Development community

Scores one gate's artifact set against that gate's rubric and writes a findings report the human reads alongside the gate review.

How to install Rubric Judge

Installs to ~/.claude/agents/tammai-product-rebuild-rubric-judge.md

Terminal

mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/tammai/product-rebuild/HEAD/agents/rubric-judge.md -o ~/.claude/agents/tammai-product-rebuild-rubric-judge.md

Restart Claude Code, or start a new session, for it to be picked up.

What Rubric Judge does

name: rubric-judge description: Scores one gate's artifact set against that gate's rubric and writes a findings report the human reads alongside the gate review. Used by the rebuild-pipeline orchestrator at phase Step 5, after validate.py passes and before the gate review is presented. Advisory only — it never gates, and it never decides. model: sonnet effort: high

You score one gate's artifacts against that gate's rubric and write a report. You do not approve anything, you do not edit

Alternatives in Development

Context Driven Development — Guide for implementing and maintaining context as a managed artifact alongside code, enabling consis 31.9k ★
After Identifying Or Executing A Chain — findings.sh add chain " " --score \ --steps " step2 -> step3>" --mitre " " findings.sh update chain --status < 213 ★
Mathodology Award Judge — Use for one independent blind judge seat scoring an award-level submission against contest rubric criteria 137 ★