compare-cache-runs — Claude Code skill: Tuist cache analysis
compare-cache-runs is a Claude Code skill that compares two Tuist cache runs, revealing hit-rate changes, root causes, and cascade effects through dependencies.
no license 291 installs ★ 35 updated 2026-03-27
npx skills add tuist/agent-skills --skill compare-cache-runs -a claude-codeWhat compare-cache-runs does
The skill takes two cache run identifiers (run ID, dashboard URL, or branch name), fetches metrics via tuist cache-run list and tuist cache-run show, and compares execution time, local and remote cache hits, and cacheable target counts.
When cache hit rate drops, the skill examines git diff to find which files changed, maps them to Tuist modules, and flags modules with direct source changes as root causes. It then calculates cascade impact: if module B depends on module A and A is invalidated, B loses cache too even though its source didn't change.
Output includes a verdict (cache improved, regressed, or stable), hit-rate delta, root cause targets, cascade breadth, and actionable recommendations like splitting large modules, using interface patterns, or pre-warming caches before merge.
What it helps with
- compare cache hit rates between two project versions
- identify which module broke cache for downstream targets
- measure build-time impact of changes through cache metrics
- distinguish expected cache misses from unexpected ones
- get recommendations to reduce cache invalidation
Example prompts for the agent
- Compare cache between main and my current feature branch
- Show me why run-123 and run-456 have different hit rates
- Analyze cache invalidation cascade from commits abc1234 to def5678
How to install compare-cache-runs
1Skills CLI (Vercel)
npx skills add tuist/agent-skills --skill compare-cache-runs -a claude-code2Manually: put the files in place
~/.claude/skills/compare-cache-runs/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/compare-cache-runs/
1Skills CLI (Vercel)
npx skills add tuist/agent-skills --skill compare-cache-runs -a codex2Manually: put the files in place
.agents/skills/compare-cache-runs/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add tuist/agent-skills --skill compare-cache-runs -a cursor2Manually: put the files in place
.cursor/skills/compare-cache-runs/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install tuist/agent-skills skills/compare-cache-runs2Manually: put the files in place
.github/skills/compare-cache-runs/SKILL.md1Skills CLI (Vercel)
npx skills add tuist/agent-skills --skill compare-cache-runs -a gemini-cli2Platform CLI
gemini skills install https://github.com/tuist/agent-skills --path skills/compare-cache-runsManually: put the files in place
.devin/skills/compare-cache-runs/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add tuist/agent-skills --skill compare-cache-runs -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about compare-cache-runs
How do I install compare-cache-runs for Claude Code?
Run npx skills add tuist/agent-skills --skill compare-cache-runs -a claude-code. The skill will be available at ~/.claude/skills/compare-cache-runs/SKILL.md and ready to use.
What does compare-cache-runs do?
It takes two Tuist cache runs, compares metrics like duration and cache hits, identifies modules with direct changes, and shows how those changes propagate through the dependency graph to affect other modules.
How does the skill determine root causes of cache invalidation?
The skill examines git diff between the two commits, maps changed files to Tuist modules, and flags modules with direct source modifications as roots. Other invalidated modules are secondary—they lost cache due to dependency changes.