trace-claude-code
# Judgeval Tracing for Claude Code
What this skill does
# Judgeval Tracing for Claude Code
Automatically trace Claude Code sessions to Judgeval for observability and debugging.
## Features
- **Session Tracing**: Capture complete conversation sessions with start/end times
- **Turn Tracking**: Track each user prompt as a separate turn
- **LLM Spans**: Log every model call with input/output and token usage
- **Tool Spans**: Track tool invocations (file operations, terminal, MCP tools)
- **Cache Metrics**: Track cache creation and read tokens for prompt caching
## Trace Structure
```
Session (task span)
├── Turn 1 (task span)
│ ├── claude-opus-4-5 (llm span)
│ ├── Read: file.py (tool span)
│ ├── Subagent: code-reviewer (task span) ← Subagent with nested spans
│ │ ├── claude-3-5-haiku (llm span)
│ │ ├── Read (tool span)
│ │ └── claude-3-5-haiku (llm span)
│ └── claude-opus-4-5 (llm span)
└── Turn 2 (task span)
└── ...
```
## Setup
After installing the plugin, run setup in your project directory:
```bash
cd /path/to/your/project
bash ~/.claude/plugins/marketplaces/judgeval-claude-plugin/skills/trace-claude-code/setup.sh
```
This will prompt you for:
- `JUDGMENT_API_KEY` - Your Judgeval API key
- `JUDGMENT_ORG_ID` - Your organization ID
- `JUDGMENT_API_URL` - API URL (default: https://api.judgmentlabs.ai)
- `JUDGEVAL_CC_PROJECT` - Project name (default: claude-code)
## Environment Variables
| Variable | Required | Description |
|----------|----------|-------------|
| `TRACE_TO_JUDGEVAL` | Yes | Set to `true` to enable tracing |
| `JUDGMENT_API_KEY` | Yes | Your Judgeval API key |
| `JUDGMENT_ORG_ID` | Yes | Your organization ID |
| `JUDGMENT_API_URL` | No | API URL (default: https://api.judgmentlabs.ai) |
| `JUDGEVAL_CC_PROJECT` | No | Project name (default: claude-code) |
| `JUDGEVAL_CC_DEBUG` | No | Set to `true` for debug logging |
## Hooks
| Hook | Trigger | Action |
|------|---------|--------|
| `session_start.sh` | Session begins | Creates root trace span |
| `user_prompt_submit.sh` | User sends prompt | Creates Turn span |
| `post_tool_use.sh` | Tool completes | Tracks tool count |
| `stop_hook.sh` | Response complete | Marks turn for finalization |
| `subagent_stop.sh` | Subagent completes | Parses subagent transcript, creates nested spans |
| `session_end.sh` | Session ends | Creates LLM/Tool spans, finalizes session |
## Span Attributes
### Session Span
- `judgment.span_kind`: "task"
- `judgment.input`: Session description
- `judgment.output`: Completion summary
- `turn_count`: Number of turns
### LLM Span
- `judgment.span_kind`: "llm"
- `judgment.input`: Conversation history
- `judgment.output`: Model response
- `judgment.llm.model`: Model name
- `judgment.llm.provider`: "anthropic"
- `judgment.usage.non_cached_input_tokens`: Input tokens
- `judgment.usage.output_tokens`: Output tokens
- `judgment.usage.cache_creation_input_tokens`: Cache write tokens
- `judgment.usage.cache_read_input_tokens`: Cache read tokens
### Tool Span
- `judgment.span_kind`: "tool"
- `judgment.input`: Tool input
- `judgment.output`: Tool output
- `tool_name`: Tool identifier
## Logs
Hook logs are written to: `~/.claude/state/judgeval_hook.log`
Enable debug logging:
```bash
export JUDGEVAL_CC_DEBUG=true
```
## Troubleshooting
**Traces not appearing:**
1. Check `TRACE_TO_JUDGEVAL=true` is set
2. Verify API key and org ID are correct
3. Check logs for errors
**Missing spans:**
1. Ensure all hooks are executable
2. Check for jq/curl availability
3. Review debug logs
Related in AI Agents
skill-development
IncludedComprehensive meta-skill for creating, managing, validating, auditing, and distributing Claude Code skills and slash commands (unified in v2.1.3+). Provides skill templates, creation workflows, validation patterns, audit checklists, naming conventions, YAML frontmatter guidance, progressive disclosure examples, and best practices lookup. Use when creating new skills, validating existing skills, auditing skill quality, understanding skill architecture, needing skill templates, learning about YAML frontmatter requirements, progressive disclosure patterns, tool restrictions (allowed-tools), skill composition, skill naming conventions, troubleshooting skill activation issues, creating custom slash commands, configuring command frontmatter, using command arguments ($ARGUMENTS, $1, $2), bash execution in commands, file references in commands, command namespacing, plugin commands, MCP slash commands, Skill tool configuration, or deciding between skills vs slash commands. Delegates to docs-management skill for official documentation.
reprompter
IncludedTransform messy prompts into well-structured, effective prompts — single or multi-agent. Use when: "reprompt", "reprompt this", "clean up this prompt", "structure my prompt", rough text needing XML tags and best practices, "reprompter teams", "repromptception", "run with quality", "smart run", "smart agents", multi-agent tasks, audits, parallel work, anything going to agent teams. Don't use when: simple Q&A, pure chat, immediate execution-only tasks. See "Don't Use When" section for details. Outputs: Structured XML/Markdown prompt, quality score (before/after), optional team brief + per-agent sub-prompts, agent team output files. Success criteria: Single mode quality score ≥ 7/10; Repromptception per-agent prompt quality score 8+/10; all required sections present, actionable and specific.
adaptive-compaction
IncludedAdaptive add-on policy and recovery layer that decides WHEN to compact, prune, snapshot, or fork -- replacing fixed-percent auto-compaction across Claude Code, Codex, and MCP-capable hosts. Trigger on auto-compact timing or damage: "when should I compact", "is it safe to compact now or start a fresh session", "auto-compact fires too early/mid-task", "switching to an unrelated task but the window still has space", "context rot", "answers get worse the longer the session runs", "the agent forgot the plan or my decisions after it summarized", "add a layer on top that manages context without changing the agent", raising autoCompactWindow to give the policy room, or installing/tuning a cross-tool compaction policy or PreCompact hook -- even when "compaction" is never said but the problem is context-window pressure or post-summarization memory loss. Do NOT use to summarize a conversation, build RAG, write a summarization prompt (decides WHEN not HOW), or answer max-context-length trivia.
agent-skill-creator
IncludedCreate cross-platform agent skills from workflow descriptions. Activates when users ask to create an agent, automate a repetitive workflow, create a custom skill, or need advanced agent creation. Triggers on phrases like create agent for, automate workflow, create skill for, every day I have to, daily I need to, turn process into agent, need to automate, create a cross-platform skill, validate this skill, export this skill, migrate this skill. Supports single skills, multi-agent suites, transcript processing, template-based creation, interactive configuration, cross-platform export, and spec validation.
llm-wiki
IncludedUse when building or maintaining a persistent personal knowledge base (second brain) in Obsidian where an LLM incrementally ingests sources, updates entity/concept pages, maintains cross-references, and keeps a synthesis current. Triggers include "second brain", "Obsidian wiki", "personal knowledge management", "ingest this paper/article/book", "build a research wiki", "compound knowledge", "Memex", or whenever the user wants knowledge to accumulate across sessions instead of being re-derived by RAG on every query.
skill-master
IncludedAgent Skills authoring, evaluation, and optimization. Create, edit, validate, benchmark, and improve skills following the agentskills.io specification. Use when designing SKILL.md files, structuring skill folders (references, scripts, assets), ingesting external documentation into skills, running trigger evals, benchmarking skill quality, optimizing descriptions, or performing blind A/B comparisons. Keywords: agentskills.io, SKILL.md, skill authoring, eval, benchmark, trigger optimization.