[shunuen0]

Copy-only companion to Judgment's Behaviors → Labels rename. Wording matches the judgment-plugin draft (JudgmentLabs/judgment-plugin#1). Noun lists say labels. A judge applies a label; the automation condition is Judge applied label. Judge type copy uses Agent judge / Code judge. Rubric scoring is an Agent judge, not a prompt scorer.

Merge this only after judgment-mono #5273 deploys. The "Labels or scores" row links to https://docs.judgmentlabs.ai/documentation/judges/monitor-judge-results, and that target goes live with #5273. https://docs.judgmentlabs.ai/documentation/monitoring is already live.

The generated skills/judgeval-jql/* files are unchanged. They are marked "Do not edit" and come with the jql follow-up, which regenerates them from the upstream jql repo. The search_traces wire field stays "behaviors". Paths, the code-judges keyword, and the live /prompt-scorers docs URL are unchanged. User feedback is captured as scores, because judges apply labels.

Before / after

• README.md: "production traces, behaviors, prompts, automations…" → "production traces, labels, prompts, automations…"
• README.md: "evaluations, code judges, datasets" → "evaluations, Code judges, datasets"
• README.md: "using code judges" → "using Code judges"
• README.md: "Create or debug Python code judges" → "Create or debug Python Code judges"
• skills/judgment/references/tracing.md: "capture as scores or behavior signals" → "capture as scores"
• skills/judgment/references/tracing.md: "Determines scoring or behavior-monitoring follow-up" → "Determines whether to follow up with a score or a label"
• skills/judgment/references/tracing.md: "Behavior or score signals | Enables quality filtering and production monitoring | …/performance/agent-behavior-monitoring" → "Labels or scores | Filter by label or score and track how often each label is applied | https://docs.judgmentlabs.ai/documentation/judges/monitor-judge-results"
• skills/judgment/references/tracing.md: "Behaviors: Filter traces or sessions by monitored behavior" → "Labels: Filter traces or sessions by the labels judges applied"
• skills/judgment/references/tracing.md: "Route failures, behaviors, or thresholds…" → "Route failures, a "Judge applied label" condition, or thresholds…"
• skills/judgment/references/code-judges.md: "# Judgment Code Judges" → "# Judgment Code judges"
• skills/judgment/references/code-judges.md: "## Current Support" → "## Current support"
• skills/judgment/references/code-judges.md: "## When To Use A Code Judge" → "## When to use a Code judge"
• skills/judgment/references/code-judges.md: "## Minimal Pattern" → "## Minimal pattern"
• skills/judgment/references/code-judges.md: "## Trace-Aware Pattern" → "## Trace-aware pattern"
• skills/judgment/references/code-judges.md: "## Packaging Guidance" → "## Packaging guidance"
• skills/judgment/references/code-judges.md: "creating Python code judges" → "creating Python Code judges"
• skills/judgment/references/code-judges.md: "Use a code judge when…" → "Use a Code judge when…"
• skills/judgment/references/agent-testing.md: "Custom code judges can also use the fields copied" → "Code judges can also use the fields copied"
• skills/judgment/references/evaluations.md: "prompt or hosted scorer for rubric-based…" → "Agent judge or hosted scorer for rubric-based…"
• skills/judgment/references/evaluations.md: "Python code judge for deterministic logic…" → "Python Code judge for deterministic logic…"
• skills/judgment/references/evaluations.md: "Use prompt scorers for qualitative rubrics…" → "Use Agent judges for qualitative rubrics…"
• skills/judgment/references/evaluations.md: "Use code judges for exact checks…" → "Use Code judges for exact checks…"
• skills/judgment/references/docs.md: "Prompt scorers:" → "Agent judges:" (URL stays …/evaluation/prompt-scorers)
• skills/judgment/SKILL.md: "evaluations, code judges, datasets" → "evaluations, Code judges, datasets"
• skills/judgment/SKILL.md: "creating code judges" → "creating Code judges"
• skills/judgment/SKILL.md: "evaluations or code judges" → "evaluations or Code judges"
• skills/judgment/SKILL.md: "Use prompt/hosted scorers for rubric-based judgment and Python code judges…" → "Use Agent judges for rubric-based judgment and Python Code judges…"
• skills/judgment/SKILL.md: "Creating Python code judges" → "Creating Python Code judges" (path references/code-judges.md unchanged)
• .claude-plugin/plugin.json: "evaluations, code judges, MCP" → "evaluations, Code judges, MCP"
• .cursor-plugin/plugin.json: "evaluations, code judges, MCP" → "evaluations, Code judges, MCP"
• .claude-plugin/marketplace.json: "evaluations, code judges, MCP" → "evaluations, Code judges, MCP"
• .claude-plugin/marketplace.json: "using code judges" → "using Code judges"
• skills/mcp-server-best-practices/SKILL.md: "traces, behaviors, sessions…" → "traces, labels, sessions…"
• skills/mcp-server-best-practices/SKILL.md: // Behaviors (any of the listed judge/value pairs) → // Labels (any of the listed judge/value pairs)

  

[nlemoff]

LGTM