Summary
Pulls MLflow documentation straight from the source by querying their llms.txt index and fetching the raw markdown files. You'd use this when working with MLflow's LLM tracking, tracing, or integrations and need accurate code examples without wading through search results or outdated Stack Overflow posts. The workflow is strict about avoiding HTML files and web search, which makes sense since MLflow's docs are well-structured and versioned. One thing to note is the llms.txt index only covers GenAI topics, so if you're working with classic ML tracking features, you might need to supplement with direct doc browsing. The emphasis on verbatim code blocks is smart since summarized examples tend to skip the details that actually matter.
Install to Claude Code
npx -y skills add mlflow/skills --skill searching-mlflow-docs --agent claude-codeInstalls into .claude/skills of the current project.
Keep your Mac awake while Claude Code and 40+ AI agents run. Sleeps when they're idle.
One time payment $9 →
Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →
Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →
On Capafy, your Skill runs online 24/7 as an agent product, and you get paid every time someone uses it.
Start earning →
Files
SKILL.mdView on GitHub
MLflow Documentation Search
Workflow
- Fetch
https://mlflow.org/docs/latest/llms.txtto find relevant page paths - Fetch the
.mdfile at the identified path - Present results with verbatim code examples
Step 1: Fetch llms.txt Index
WebFetch(
url: "https://mlflow.org/docs/latest/llms.txt",
prompt: "Find links or references to [TOPIC]. List all relevant URLs."
)
Step 2: Fetch Target Documentation
Use the path from Step 1, always with .md extension:
WebFetch(
url: "https://mlflow.org/docs/latest/[path].md",
prompt: "Return all code blocks verbatim. Do not summarize."
)
Anti-Patterns
Do not use .html files — Fetch .md source files only.
Do not use WebSearch — Always start from llms.txt; web search returns outdated or third-party content.
Do not use vague prompts — "Extract complete documentation" allows summarization. Use "Return all code blocks verbatim. Do not summarize."
Do not use versioned paths — Always use /docs/latest/, never /docs/3.8/ or other versions unless the user explicitly requests a specific version.
Do not guess URLs — Always verify paths exist in llms.txt before fetching. Never construct documentation paths from assumptions.
Do not follow external links — Stay within mlflow.org/docs. Do not follow links to GitHub, PyPI, or third-party sites.
Do not mix sources — Use only MLflow docs. Do not combine with LangChain docs, OpenAI docs, or other external documentation.
Do not use llms.txt for non-GenAI topics — The llms.txt index covers LLM/GenAI documentation only. For classic ML tracking features, paths may differ.
Featured
Keep your Mac awake while Claude Code and 40+ AI agents run. Sleeps when they're idle.
One time payment $9 →Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →On Capafy, your Skill runs online 24/7 as an agent product, and you get paid every time someone uses it.
Start earning →
