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

retrieving-mlflow-traces

Retrieves and filters MLflow traces by ID, status, session, tags, and execution time via CLI or Python API.

Updated Mar 5, 2026
Search Tags:#debugging#cli#search#traces#filter#mlflow#llm observability
One-click install
npx skills add https://github.com/FMurray/mlfts --skill retrieving-mlflow-traces-fmurray
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: retrieving-mlflow-traces
Source: https://github.com/FMurray/mlfts/tree/main/.agents/skills/retrieving-mlflow-traces
Command: npx skills add https://github.com/FMurray/mlfts --skill retrieving-mlflow-traces-fmurray

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires mlflow.

What problem does it solve? Debugging LLM applications requires finding specific traces among thousands, but constructing correct MLflow filter syntax for sessions, statuses, timestamps, and metadata is error-prone and poorly documented. ## Core Features & Use Cases - Single Trace Fetch: Retrieve full trace details by trace ID using mlflow traces get. - Filtered Search: Query traces by run ID, session, user, status, time range, execution time, tags, metadata, and assessments using mlflow traces search with filter-string syntax. - Pagination & Output Control: Handle large result sets with page tokens, ordering, JSON output, and optional span inclusion. - Use Case: A user reports a failed agent session. Search traces by mlflow.trace.session metadata combined with trace.status = 'ERROR', ordered by timestamp, to reconstruct the sequence of events and find the root cause. ## Quick Start Ask the assistant to find all failed MLflow traces from a specific session in your experiment and show their details.

Frequently Asked Questions about retrieving-mlflow-traces

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

FAQPage Schema
How do I search MLflow traces by session ID?▼

Use mlflow traces search with a filter string on the session metadata key: metadata.`mlflow.trace.session` = 'session_id'. Backticks are required around the key because it contains dots, and you must pass the experiment ID with --experiment-id.

How to find failed or slow traces in MLflow?▼

Filter by trace.status = 'ERROR' to find failed traces, or trace.execution_time_ms > 1000 to find traces slower than one second. Combine both conditions with AND in a single filter string for targeted debugging.

What is the difference between mlflow traces get and search?▼

Use mlflow traces get --trace-id when you already have a specific trace ID from the UI or logs. Use mlflow traces search when you need to find traces by criteria like session, user, status, or time range.

Why does my MLflow trace filter string fail with special characters?▼

Metadata or tag keys containing dots or dashes must be wrapped in backticks, such as tag.`model-name` or metadata.`user.id`. Without backtick escaping, the filter parser misinterprets the key structure and the query fails.

How do I paginate through large MLflow trace search results?▼

Use --max-results to limit results per page. When more results exist, the output includes a next page token, which you pass with --page-token to fetch the subsequent page of traces.

Related Skills

Explore More Software Engineering Skills→
JingyiWngJingyiWng

retrieving-mlflow-traces

Retrieve MLflow traces by ID or advanced filters across experiments.

Community
Advanced
databricks-solutionsdatabricks-solutions
5

mlflow-genai-evaluation

Evaluate Databricks GenAI agents with MLflow GenAI LLM-as-judge scorers.

Official
Advanced
yanochka11yanochka11

build-review-interface

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

Community
Advanced
biomaps-infrabiomaps-infra

service-labarchive-integration

Integrate with the LabArchives API to manage notebooks, entries, and attachments.

Official
Advanced