Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Package entire code repositories into single AI-friendly files using Repomix. Capabilities include pack codebases with customizable include/exclude patterns, generate multiple output formats (XML, Markdown, plain text), preserve file structure and context, optimize for AI consumption with token counting, filter by file types and directories, add custom headers and summaries. Use when packaging codebases for AI analysis, creating repository snapshots for LLM context, analyzing third-party librari
.claude/skills/microck-repomix/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-05 | ✗→✓ | ▲ Improved | 42% | 0% |
| case-18 | ✓→✓ | ＝ Same ✓ | 0% | 0% |
| case-01 | ✓→✓ | ＝ Same ✓ | 54% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | 159% | 0% |
| case-03 | ✓→✓ | ＝ Same ✓ | 171% | 0% |
Repomix packs entire repositories into single, AI-friendly files. Perfect for feeding codebases to LLMs like Claude, ChatGPT, and Gemini.
Use when:
bashrepomix --version
bash# npm npm install -g repomix # Homebrew (macOS/Linux) brew install repomix
bash# Package current directory (generates repomix-output.xml) repomix # Specify output format repomix --style markdown repomix --style json # Package remote repository npx repomix --remote owner/repo # Custom output with filters repomix --include "src/**/*.ts" --remove-comments -o output.md
Process remote repositories without cloning:
bash# Shorthand npx repomix --remote yamadashy/repomix # Full URL npx repomix --remote https://github.com/owner/repo # Specific commit npx repomix --remote https://github.com/owner/repo/commit/hash
Strip comments from supported languages (HTML, CSS, JavaScript, TypeScript, Vue, Svelte, Python, PHP, Ruby, C, C#, Java, Go, Rust, Swift, Kotlin, Dart, Shell, YAML):
bashrepomix --remove-comments
bash# Package feature branch for AI review repomix --include "src/**/*.ts" --remove-comments -o review.md --style markdown
bash# Package third-party library npx repomix --remote vendor/library --style xml -o audit.xml
bash# Package with docs and code repomix --include "src/**,docs/**,*.md" --style markdown -o context.md
bash# Package specific modules repomix --include "src/auth/**,src/api/**" -o debug-context.xml
bash# Full codebase context repomix --remove-comments --copy
bash# Include specific patterns repomix --include "src/**/*.ts,*.md" # Ignore additional patterns repomix -i "tests/**,*.test.js" # Disable .gitignore rules repomix --no-gitignore
bash# Output format repomix --style markdown # or xml, json, plain # Output file path repomix -o output.md # Remove comments repomix --remove-comments # Copy to clipboard repomix --copy
bash# Use custom config file repomix -c custom-config.json # Initialize new config repomix --init # creates repomix.config.json
Repomix automatically counts tokens for individual files, total repository, and per-format output.
Typical LLM context limits:
Repomix uses Secretlint to detect sensitive data (API keys, passwords, credentials, private keys, AWS secrets).
Best practices:
.repomixignore for sensitive files.env filesDisable security checks if needed:
bashrepomix --no-security-check
When user requests repository packaging:
For detailed information, see:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-18 | pass→pass | 14,163 | 5,893 | -58% | 1 | 1 | 0% | 2,311 | 2,301 | -0% | 0 | 0 | — |
case-01 | pass→pass | 6,405 | 4,083 | -36% | 1 | 1 | 0% | 1,162 | 1,790 | +54% | 0 | 0 | — |
case-02 | pass→pass | 3,761 | 2,963 | -21% | 1 | 1 | 0% | 648 | 1,681 | +159% | 0 | 0 | — |
case-03 | pass→pass | 4,353 | 3,430 | -21% | 1 | 1 | 0% | 706 | 1,915 | +171% | 0 | 0 | — |
case-04 | pass→pass | 4,183 | 2,938 | -30% | 1 | 1 | 0% | 654 | 1,724 | +164% | 0 | 0 | — |
case-05 | fail→pass | 6,829 | 2,932 | -57% | 1 | 1 | 0% | 1,259 | 1,783 | +42% | 0 | 0 | — |
case-06 | pass→pass | 7,329 | 3,482 | -52% | 1 | 1 | 0% | 1,273 | 1,861 | +46% | 0 | 0 | — |
case-07 | pass→pass | 2,934 | 1,610 | -45% | 1 | 1 | 0% | 496 | 1,563 | +215% | 0 | 0 | — |
case-08 | pass→pass | 7,526 | 3,038 | -60% | 1 | 1 | 0% | 1,443 | 1,774 | +23% | 0 | 0 | — |
case-09 | pass→pass | 3,315 | 2,376 | -28% | 1 | 1 | 0% | 394 | 1,635 | +315% | 0 | 0 | — |
case-10 | pass→pass | 3,688 | 4,260 | +16% | 1 | 1 | 0% | 547 | 1,817 | +232% | 0 | 0 | — |
case-11 | pass→pass | 2,957 | 2,479 | -16% | 1 | 1 | 0% | 436 | 1,452 | +233% | 0 | 0 | — |
case-12 | pass→pass | 2,565 | 1,785 | -30% | 1 | 1 | 0% | 474 | 1,566 | +230% | 0 | 0 | — |
case-13 | pass→pass | 4,990 | 3,469 | -30% | 1 | 1 | 0% | 947 | 1,854 | +96% | 0 | 0 | — |
case-14 | pass→pass | 7,479 | 2,112 | -72% | 1 | 1 | 0% | 1,195 | 1,578 | +32% | 0 | 0 | — |
case-15 | pass→pass | 9,206 | 3,426 | -63% | 1 | 1 | 0% | 1,558 | 1,782 | +14% | 0 | 0 | — |
case-16 | pass→pass | 2,906 | 1,594 | -45% | 1 | 1 | 0% | 468 | 1,454 | +211% | 0 | 0 | — |
case-17 | pass→pass | 6,095 | 3,606 | -41% | 1 | 1 | 0% | 1,125 | 1,747 | +55% | 0 | 0 | — |
case-19 | pass→pass | 7,560 | 3,875 | -49% | 1 | 1 | 0% | 1,284 | 1,841 | +43% | 0 | 0 | — |
case-20 | pass→pass | 10,701 | 6,133 | -43% | 1 | 1 | 0% | 1,896 | 2,403 | +27% | 0 | 0 | — |
case-21 | pass→pass | 8,347 | 7,090 | -15% | 1 | 1 | 0% | 1,530 | 2,620 | +71% | 0 | 0 | — |
case-22 | pass→pass | 8,863 | 7,310 | -18% | 1 | 1 | 0% | 1,681 | 2,673 | +59% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +5 percentage points is the difference between those two pass rates over the 22 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.