skills.rest
FMurray
FMurrayCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

tag

Adds key-value tags to MLflow traces for filtering and session metadata.

Updated Mar 5, 2026
Search Tags:#claude code#metadata#annotation#mlflow#databricks#llm observability#trace tagging
One-click install
npx skills add https://github.com/FMurray/mlfts --skill tag-fmurray
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: tag
Source: https://github.com/FMurray/mlfts/tree/main/.claude/skills/tag-trace
Command: npx skills add https://github.com/FMurray/mlfts --skill tag-fmurray

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires mlflow, and includes scripts (resource) components.

What problem does it solve? After MLflow traces are logged, there is no quick way to attach labels or metadata from within a Claude Code session, making it hard to filter, organize, and review traces later. ## Core Features & Use Cases - Trace Tagging: Attach one or more key=value tags to the most recent trace or a specific trace ID via the trace_annotate.py script. - Trace Listing: List recent traces with ID, timestamp, status, and input preview to find the trace you want to annotate. - Use Case: After a coding session, tag the resulting trace with quality=good and sprint=42 so you can later filter traces in MLflow with search queries like tag.quality = 'good'. ## Quick Start Ask the assistant to tag the most recent MLflow trace with quality=good using the tag skill.

Frequently Asked Questions about tag

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

FAQPage Schema
How do I tag an MLflow trace after it is logged?▼

Run the trace_annotate.py script with the tag subcommand and one or more key=value pairs. By default it tags the most recent trace, or you can pass --trace-id to target a specific trace.

How do I find the trace ID of a recent MLflow trace?▼

Use the list subcommand of trace_annotate.py to show recent traces with their IDs, timestamps, status, and input previews. The --max-results flag controls how many traces are displayed.

Can I filter MLflow traces by tags?▼

Yes, tags are stored on the trace's info.tags dictionary and can be used in search filters. For example, search with a filter string like tag.quality = 'good' to find matching traces.

What happens if no traces exist when tagging?▼

The script exits with an error message stating no traces were found in the current experiment. It also excludes env_snapshot companion traces when selecting the most recent trace.

Does the tag skill require a specific MLflow tracking setup?▼

Yes, it calls setup_mlflow from mlflow.claude_code.tracing and uses the configured tracking URI, typically Databricks. The tracking URI must be set via environment variables such as MLFLOW_TRACKING_URI before running.

Related Skills

Explore More Software Engineering Skills→
mlflowmlflow
27.8k

status

Displays the current MLflow tracing configuration for Claude Code.

Official
Basic
JingyiWngJingyiWng

databricks-mlflow-evaluation

Evaluate MLflow GenAI agents with built-in and custom metrics.

Community
Advanced
yanochka11yanochka11

build-review-interface

Build browser-based annotation interfaces for reviewing LLM traces with pass/fail labels.

Community
Advanced
Aradhya0510Aradhya0510
3

retrieving-mlflow-traces

Retrieve and search MLflow traces by ID, status, session, user, tags, metadata, execution time, and time range.

Community
Intermediate