Spark Advisor
Turns Spark History Server metrics into severity-rated skew/GC/shuffle findings with exact thresholds.
Test report
- Verdict
- Tested · Works
- Score
- Tested
- Jul 14, 2026
- Environment
- Claude Code 2.x (agent harness)
- Upstream re-checked
- Aug 10, 2026 · caafacb
pip install spark-history-cli succeeded for real and the CLI runs, but this sandbox has no Java/Spark to stand up a real History Server, so the diagnostic thresholds (p95/p50>10x skew, GC>20%, shuffle>2x input) were verified against a described metrics scenario rather than a live pull — evidence, not faith. Worth flagging: the repo ships two SKILL.md copies (skills/spark-advisor and spark_history_cli/skills/spark-advisor) and the packaged one is stale, missing the Quick Start and SQL Plan Analysis sections present in the canonical top-level copy used for install.
Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works
- Installs cleanly 5/5
- Triggers reliably 5/5
- Output vs. baseline 7/10
- Docs & honesty 4/5
What Spark Advisor does
Spark Advisor is a Claude Code skill that drives the spark-history-cli CLI against a running Spark History Server to diagnose slow Spark apps and SQL queries, compare TPC-DS/benchmark runs, and produce tuning recommendations. Trigger it when the user asks why a Spark job or query is slow, wants to compare two runs, or asks about skew, GC pressure, shuffle overhead, spill, or Gluten/Iceberg/Delta-specific issues.
How to install Spark Advisor
git clone https://github.com/yaooqinn/spark-history-cli
cd spark-history-cli
mkdir -p ~/.claude/skills
cp -r skills/spark-advisor ~/.claude/skills/spark-advisor
Skills live in ~/.claude/skills/ (global) or .claude/skills/
(per-project). Restart Claude Code after installing.
Commands — how to trigger Spark Advisor
-
/spark-advisorTurns Spark History Server metrics into severity-rated skew/GC/shuffle findings with exact thresholds.
It also activates on plain-language prompts like these:
-
My Spark job is slow, here's the history server metrics for it -
This TPC-DS query has bad stage skew, can you diagnose it? -
GC time is eating my Spark job, help me tune the executors
Frequently asked questions
- Is the Spark Advisor skill free?
- Yes. The skill itself is free from yaooqinn/spark-history-cli. SkillProof publishes the install command and an independent test verdict at no cost.
- Does Spark Advisor work with Claude Code?
- We tested it with Claude Code 2.x (agent harness) on Jul 14, 2026. Verdict: Tested · Works. pip install spark-history-cli succeeded for real and the CLI runs, but this sandbox has no Java/Spark to stand up a real History Server, so the diagnostic thresholds (p95/p50>10x skew, GC>20%, shuffle>2x input) were verified against a described metrics scenario rather than a live pull — evidence, not faith. Worth flagging: the repo ships two SKILL.md copies (skills/spark-advisor and spark_history_cli/skills/spark-advisor) and the packaged one is stale, missing the Quick Start and SQL Plan Analysis sections present in the canonical top-level copy used for install.
- What is the Spark Advisor SkillProof Score?
- 8.4/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 7/10, docs & honesty 4/5.
- How do I install Spark Advisor?
- Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
- Can I use Spark Advisor with Cursor, Copilot, Gemini CLI, Codex or other AI tools?
- The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.