repomix

Codex compatibility note:

• Invoke repository skills with $skill-name in Codex; this mirrored copy rewrites legacy Claude /skill-name references.
• Task tracker mandate: BEFORE executing any workflow or skill step, create/update task tracking for all steps and keep it synchronized as progress changes.
• User-question prompts mean to ask the user directly in Codex.
• Ignore Claude-specific mode-switch instructions when they appear.
• Strict execution contract: when a user explicitly invokes a skill, execute that skill protocol as written.
• Subagent authorization: when a skill is user-invoked or AI-detected and its protocol requires subagents, that skill activation authorizes use of the required spawn_agent subagent(s) for that task.
• Do not skip, reorder, or merge protocol steps unless the user explicitly approves the deviation first.
• For workflow skills, execute each listed child-skill step explicitly and report step-by-step evidence.
• If a required step/tool cannot run in this environment, stop and ask the user before adapting.

Codex Project-Reference Loading (No Hooks)

Codex does not receive Claude hook-based doc injection. When coding, planning, debugging, testing, or reviewing, open project docs explicitly using this routing.

Always read:

• docs/project-config.json (project-specific paths, commands, modules, and workflow/test settings)
• docs/project-reference/docs-index-reference.md (routes to the full docs/project-reference/* catalog)
• docs/project-reference/lessons.md (always-on guardrails and anti-patterns)

Situation-based docs:

• Backend/CQRS/API/domain/entity changes: backend-patterns-reference.md, domain-entities-reference.md, project-structure-reference.md
• Frontend/UI/styling/design-system: frontend-patterns-reference.md, scss-styling-guide.md, design-system/README.md
• Spec/test-case planning or TC mapping: feature-docs-reference.md
• Integration test implementation/review: integration-test-reference.md
• E2E test implementation/review: e2e-test-reference.md
• Code review/audit work: code-review-rules.md plus domain docs above based on changed files

Do not read all docs blindly. Start from docs-index-reference.md, then open only relevant files for the task.

Quick Summary

Goal: Package code repositories into single AI-friendly files using Repomix for LLM analysis.

Workflow:

1. Assess — Identify target repo (local/remote), output format, sensitivity concerns
2. Configure — Set include/ignore patterns, comment removal, output style
3. Execute — Run repomix with options, monitor token counts
4. Validate — Review output for sensitive data, check token limits for target LLM

Key Rules:

• Always review output before sharing (security check for API keys, credentials)
• Use --include patterns to stay within LLM context limits
• Default output is XML; use --style markdown for readable output
• Use --token-count-tree to identify token-heavy files before packaging

Be skeptical. Apply critical thinking, sequential thinking. Every claim needs traced proof, confidence percentages (Idea should be more than 80%).

Repomix Skill

Repomix packs entire repositories into single, AI-friendly files. Perfect for feeding codebases to LLMs like Claude, ChatGPT, and Gemini.

When to Use

Use when:

• Packaging codebases for AI analysis
• Creating repository snapshots for LLM context
• Analyzing third-party libraries
• Preparing for security audits
• Generating documentation context
• Investigating bugs across large codebases
• Creating AI-friendly code representations

Quick Start

Check Installation

repomix --version

Install

# npm
npm install -g repomix

# Homebrew (macOS/Linux)
brew install repomix

Basic Usage

# 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

Core Capabilities

Repository Packaging

• AI-optimized formatting with clear separators
• Multiple output formats: XML, Markdown, JSON, Plain text
• Git-aware processing (respects .gitignore)
• Token counting for LLM context management
• Security checks for sensitive information

Remote Repository Support

Process remote repositories without cloning:

# 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

Comment Removal

Strip comments from supported languages (HTML, CSS, JavaScript, TypeScript, Vue, Svelte, Python, PHP, Ruby, C, C#, Java, Go, Rust, Swift, Kotlin, Dart, Shell, YAML):

repomix --remove-comments

Common Use Cases

Code Review Preparation

# Package feature branch for AI review
repomix --include "src/**/*.ts" --remove-comments -o review.md --style markdown

Security Audit

# Package third-party library
npx repomix --remote vendor/library --style xml -o audit.xml

Documentation Generation

# Package with docs and code
repomix --include "src/**,docs/**,*.md" --style markdown -o context.md

Bug Investigation

# Package specific modules
repomix --include "src/auth/**,src/api/**" -o debug-context.xml

Implementation Planning

# Full codebase context
repomix --remove-comments --copy

Command Line Reference

File Selection

# Include specific patterns
repomix --include "src/**/*.ts,*.md"

# Ignore additional patterns
repomix -i "tests/**,*.test.js"

# Disable .gitignore rules
repomix --no-gitignore

Output Options

# 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

Configuration

# Use custom config file
repomix -c custom-config.json

# Initialize new config
repomix --init  # creates repomix.config.json

Token Management

Repomix automatically counts tokens for individual files, total repository, and per-format output.

Typical LLM context limits:

• Claude Sonnet 4.5: ~200K tokens
• GPT-4: ~128K tokens
• GPT-3.5: ~16K tokens

Token Count Optimization

Understanding your codebase's token distribution is crucial for optimizing AI interactions. Use the --token-count-tree option to visualize token usage across your project:

repomix --token-count-tree

This displays a hierarchical view of your codebase with token counts:

🔢 Token Count Tree:
────────────────────
└── src/ (70,925 tokens)
    ├── cli/ (12,714 tokens)
    │   ├── actions/ (7,546 tokens)
    │   └── reporters/ (990 tokens)
    └── core/ (41,600 tokens)
        ├── file/ (10,098 tokens)
        └── output/ (5,808 tokens)

You can also set a minimum token threshold to focus on larger files:

repomix --token-count-tree 1000  # Only show files/directories with 1000+ tokens

This helps you:

• Identify token-heavy files that might exceed AI context limits
• Optimize file selection using --include and --ignore patterns
• Plan compression strategies by targeting the largest contributors
• Balance content vs. context when preparing code for AI analysis

Security Considerations

Repomix uses Secretlint to detect sensitive data (API keys, passwords, credentials, private keys, AWS secrets).

Best practices:

1. Always review output before sharing
2. Use .repomixignore for sensitive files
3. Enable security checks for unknown codebases
4. Exclude .env files from packaging
5. Check for hardcoded credentials

Disable security checks if needed:

repomix --no-security-check

Implementation Workflow

When user requests repository packaging:

1. Assess Requirements

   • Identify target repository (local/remote)
   • Determine output format needed
   • Check for sensitive data concerns

2. Configure Filters

   • Set include patterns for relevant files
   • Add ignore patterns for unnecessary files
   • Enable/disable comment removal

3. Execute Packaging

   • Run repomix with appropriate options
   • Monitor token counts
   • Verify security checks

4. Validate Output

   • Review generated file
   • Confirm no sensitive data
   • Check token limits for target LLM

5. Deliver Context

   • Provide packaged file to user
   • Include token count summary
   • Note any warnings or issues

Reference Documentation

For detailed information, see:

• Configuration Reference - Config files, include/exclude patterns, output formats, advanced options
• Usage Patterns - AI analysis workflows, security audit preparation, documentation generation, library evaluation

Additional Resources

• GitHub: https://github.com/yamadashy/repomix
• Documentation: https://repomix.com/guide/
• MCP Server: Available for AI assistant integration

---

[IMPORTANT] Use task tracking to break ALL work into small tasks BEFORE starting — including tasks for each file read. This prevents context loss from long files. For simple tasks, AI MUST ATTENTION ask user whether to skip.

AI Mistake Prevention — Failure modes to avoid on every task: Check downstream references before deleting. Deleting components causes documentation and code staleness cascades. Map all referencing files before removal. Verify AI-generated content against actual code. AI hallucinates APIs, class names, and method signatures. Always grep to confirm existence before documenting or referencing. Trace full dependency chain after edits. Changing a definition misses downstream variables and consumers derived from it. Always trace the full chain. Trace ALL code paths when verifying correctness. Confirming code exists is not confirming it executes. Always trace early exits, error branches, and conditional skips — not just happy path. When debugging, ask "whose responsibility?" before fixing. Trace whether bug is in caller (wrong data) or callee (wrong handling). Fix at responsible layer — never patch symptom site. Assume existing values are intentional — ask WHY before changing. Before changing any constant, limit, flag, or pattern: read comments, check git blame, examine surrounding code. Verify ALL affected outputs, not just the first. Changes touching multiple stacks require verifying EVERY output. One green check is not all green checks. Holistic-first debugging — resist nearest-attention trap. When investigating any failure, list EVERY precondition first (config, env vars, DB names, endpoints, DI registrations, data preconditions), then verify each against evidence before forming any code-layer hypothesis. Surgical changes — apply the diff test. Bug fix: every changed line must trace directly to the bug. Don't restyle or improve adjacent code. Enhancement task: implement improvements AND announce them explicitly. Surface ambiguity before coding — don't pick silently. If request has multiple interpretations, present each with effort estimate and ask. Never assume all-records, file-based, or more complex path.

Critical Thinking Mindset — Apply critical thinking, sequential thinking. Every claim needs traced proof, confidence >80% to act. Anti-hallucination: Never present guess as fact — cite sources for every claim, admit uncertainty freely, self-check output for errors, cross-reference independently, stay skeptical of own confidence — certainty without evidence root of all hallucination.

MUST ATTENTION apply critical thinking — every claim needs traced proof, confidence >80% to act. Anti-hallucination: never present guess as fact.

MUST ATTENTION apply AI mistake prevention — holistic-first debugging, fix at responsible layer, surface ambiguity before coding, re-read files after compaction.

Closing Reminders

IMPORTANT MUST ATTENTION break work into small todo tasks using task tracking BEFORE starting IMPORTANT MUST ATTENTION search codebase for 3+ similar patterns before creating new code IMPORTANT MUST ATTENTION cite file:line evidence for every claim (confidence >80% to act) IMPORTANT MUST ATTENTION add a final review todo task to verify work quality

[TASK-PLANNING] Before acting, analyze task scope and systematically break it into small todo tasks and sub-tasks using task tracking.

Hookless Prompt Protocol Mirror (Auto-Synced)

Source: .claude/hooks/lib/prompt-injections.cjs + .claude/.ck.json

[WORKFLOW-EXECUTION-PROTOCOL] [BLOCKING] Workflow Execution Protocol — MANDATORY IMPORTANT MUST CRITICAL. Do not skip for any reason.

Generic portability boundary: Reusable skills and protocol text stay project-neutral; project-specific conventions are discovered from docs/project-config.json and docs/project-reference/. Apply shared AI-SDD from shared/sdd-artifact-contract.md. Read docs/project-config.json and docs/project-reference/docs-index-reference.md, then open the project reference docs named there. Any supported AI tool may execute when this shared context and local docs are available.

1. DETECT: Match prompt against workflow catalog
2. ANALYZE: Find best-match workflow AND evaluate if a custom step combination would fit better
3. ASK (REQUIRED FORMAT): Use a direct user question with this structure unless the user explicitly invoked a workflow/skill and the local protocol treats explicit invocation as confirmation:
   • Question: "Which workflow do you want to activate?"
   • Option 1: "Activate [BestMatch Workflow] (Recommended)"
   • Option 2: "Activate custom workflow: [step1 → step2 → ...]" (include one-line rationale)
4. ACTIVATE (if confirmed): Call $workflow-start <workflowId> for standard; sequence custom steps manually
5. CREATE TASKS: task tracking for ALL workflow steps
6. EXECUTE: Follow each step in sequence [CRITICAL-THINKING-MINDSET] Apply critical thinking, sequential thinking. Every claim needs traced proof, confidence >80% to act. Anti-hallucination principle: Never present guess as fact — cite sources for every claim, admit uncertainty freely, self-check output for errors, cross-reference independently, stay skeptical of own confidence — certainty without evidence root of all hallucination. AI Attention principle (Primacy-Recency): Put the 3 most critical rules at both top and bottom of long prompts/protocols so instruction adherence survives long context windows. Goal-driven execution: Define success criteria first, loop until verified, and stop only when observable checks pass. Tests verify intent: Tests must protect business rules/invariants and fail when the protected intent breaks, not only mirror current behavior.

[LESSON-LEARNED-REMINDER] [BLOCKING] Task Planning & Continuous Improvement — MANDATORY. Do not skip.

Break work into small tasks (task tracking) before starting. Add final task: "Analyze AI mistakes & lessons learned".

Extract lessons — ROOT CAUSE ONLY, not symptom fixes:

1. Name the FAILURE MODE (reasoning/assumption failure), not symptom — "assumed API existed without reading source" not "used wrong enum value".
2. Generality test: does this failure mode apply to ≥3 contexts/codebases? If not, abstract one level up.
3. Write as a universal rule — strip project-specific names/paths/classes. Useful on any codebase.
4. Consolidate: multiple mistakes sharing one failure mode → ONE lesson.
5. Recurrence gate: "Would this recur in future session WITHOUT this reminder?" — No → skip $learn.
6. Auto-fix gate: "Could $code-review/$code-simplifier/$security/$lint catch this?" — Yes → improve review skill instead.
7. BOTH gates pass → ask user to run $learn. [TASK-PLANNING] [MANDATORY] BEFORE executing any workflow or skill step, create/update task tracking for all planned steps, then keep it synchronized as each step starts/completes.