Claude Code Unveiled

The first architecture deep-dive, hidden feature catalog, and cost optimization guide for Claude Code — based on v2.1.88 source analysis (1,884 files, 132K lines of TypeScript).

中文版 | English

On March 31, 2026, Claude Code's TypeScript source (1,884 files, ~132K lines) was exposed through an unstripped .map file in the npm package. This repository includes the complete restored source code in src/ plus original analysis. This project extracts actionable insights — architecture patterns, hidden features, cost optimization, and best practices — all with source file references.

Table of Contents

• System Prompt (Complete Reconstruction)
• 87 Hidden Feature Flags
• 15 Hidden Slash Commands
• 25 Internal-Only Commands
• Undercover Mode (How Anthropic Hides AI Attribution)
• Cost Optimization (10 Tips from Source)
• Architecture Diagrams
• Claude Code vs Cursor vs Cline
• Telemetry & Privacy
• Remote Control & Killswitches
• Future Roadmap (Unreleased Features)
• CLAUDE.md Best Practices

System Prompt (Complete Reconstruction)

Full document | Source: src/constants/prompts.ts, src/utils/systemPrompt.ts

5-Layer Priority System

Priority 0: Override (loop mode, testing)        ← highest
Priority 1: Coordinator (multi-worker orchestration)
Priority 2: Agent (subagent definitions)
Priority 3: Custom (--system-prompt flag)
Priority 4: Default (standard Claude Code)       ← lowest
         + appendSystemPrompt (always appended unless override)

What the Default Prompt Actually Says

The system prompt is 9 major sections:

1. Identity — "You are an interactive agent that helps users with software engineering tasks"
2. System Rules — Tool execution, permission modes, prompt injection detection
3. Task Execution — The most detailed section:
   • Read code before modifying (mandatory)
   • Don't add unrequested features/refactoring
   • Don't create helpers for one-time operations
   • "Three similar lines of code is better than a premature abstraction"
   • Avoid OWASP Top 10 vulnerabilities
4. Careful Actions — Destructive operations need user confirmation (rm -rf, force push, etc.)
5. Tool Usage — Use dedicated tools (Read/Edit/Glob/Grep) instead of Bash
6. Tone — No emojis, concise, file_path:line_number format
7. Output Efficiency — "If you can say it in one sentence, don't use three"
8. Cache Boundary — __SYSTEM_PROMPT_DYNAMIC_BOUNDARY__ splits static/dynamic parts
9. Environment — CWD, platform, model name, knowledge cutoff

Cache Optimization

The prompt is split by __SYSTEM_PROMPT_DYNAMIC_BOUNDARY__:

• Before boundary (static): Cached globally, saves tokens on repeat calls
• After boundary (dynamic): User/session-specific, recalculated each time (memory, MCP, language, etc.)

Model Knowledge Cutoffs

87 Hidden Feature Flags

Full document | Source: src/commands.ts, various

These are compile-time switches via bun:bundle feature(). Most are dead-code-eliminated in the public npm build.

Core Codenames

Context & Compression

Obfuscated Runtime Config (tengu_* gates)

Runtime flags use tengu_ + random word pairs to deliberately hide their purpose:

tengu_frond_boric       → Analytics killswitch (Datadog/FirstParty)
tengu_passport_quail    → Memory extraction gate
tengu_moth_copse        → Memory extraction enable
tengu_bramble_lintel    → Memory extraction frequency
tengu_cicada_nap_ms     → Background refresh throttle
tengu_slate_prism       → Connector text summarization
tengu_amber_json_tools  → JSON tool format (token-efficient)
tengu_tool_pear         → Structured output (strict tools)

15 Hidden Slash Commands

Source: src/commands.ts

These require specific feature flags to appear:

25 Internal-Only Commands

Condition: USER_TYPE === 'ant' && !IS_DEMO

Undercover Mode

Full document | Source: src/utils/undercover.ts

When Anthropic employees (USER_TYPE=ant) work in public repositories, Claude Code automatically hides all internal information.

How It Works

isUndercover() check:
  USER_TYPE !== 'ant' → always OFF (external users unaffected)
  USER_TYPE === 'ant' →
    CLAUDE_CODE_UNDERCOVER=1 → forced ON
    Repo in internal allowlist → OFF
    Repo is public/unknown → ON (safe default)

There is NO force-OFF option. If Claude Code can't confirm it's an internal repo, undercover stays ON.

What Gets Injected

## UNDERCOVER MODE — CRITICAL

You are operating UNDERCOVER in a PUBLIC/OPEN-SOURCE repository.
NEVER include in commit messages or PR descriptions:
- Internal model codenames (Capybara, Tengu, etc.)
- Unreleased model version numbers
- Internal repo or project names
- The phrase "Claude Code" or any mention that you are an AI
- Co-Authored-By lines or any other attribution

Internal Repo Allowlist (22 private repos)

anthropics/claude-cli-internal, anthropics/anthropic, anthropics/apps,
anthropics/casino, anthropics/dbt, anthropics/dotfiles,
anthropics/terraform-config, anthropics/hex-export, anthropics/feedback-v2,
anthropics/labs, anthropics/argo-rollouts, anthropics/starling-configs,
anthropics/ts-tools, anthropics/ts-capsules, anthropics/feldspar-testing,
anthropics/trellis, anthropics/claude-for-hiring, anthropics/forge-web,
anthropics/infra-manifests, anthropics/mycro_manifests,
anthropics/mycro_configs, anthropics/mobile-apps

Source: src/utils/commitAttribution.ts lines 30-75

Cost Optimization

Full document | Sources: src/utils/modelCost.ts, src/utils/context.ts, src/services/compact/

Model Pricing (per million tokens)

Fast mode is 6x the price. Use it only for short, time-sensitive tasks.

Source: src/utils/modelCost.ts lines 36-69

10 Money-Saving Tips

Key Thresholds

Context window:        200,000 tokens
Auto-compact trigger:  ~167,000 tokens (200K - 20K output - 13K buffer)
Manual compact buffer:   3,000 tokens
Default max_output:      8,000 tokens (escalates to 64,000 on retry)
Session memory min:     10,000 tokens
Session memory max:     40,000 tokens
Post-compact restore:   50,000 tokens budget, 5 files max

Architecture Diagrams

Request Lifecycle

Full document

sequenceDiagram
    participant User
    participant REPL
    participant Query as QueryEngine
    participant API as Claude API
    participant Tool as Tool Executor

    User->>REPL: Input message
    REPL->>Query: submitMessage()
    Query->>Query: Build system prompt (5-layer priority)
    Query->>Query: Load CLAUDE.md + Memory
    Query->>API: messages.create(stream)

    loop Tool Loop
        API-->>Query: tool_use response
        Query->>Tool: Permission check → Execute
        Tool-->>Query: Tool result
        Query->>API: Continue with result
    end

    API-->>REPL: Final text response
    Query->>Query: Cost tracking + auto-compact check

Tool System

Full document

40+ built-in tools across 4 permission levels:

Permission Model

Full document

4 modes: Default (ask) → Plan (read-only) → Auto (smart judge) → Bypass (allow all)

Multi-Agent Architecture

Full document

Main Thread → Agent tool → Subagent (fresh context, isolated)
                        → Subagent (can use different model)
                        → Subagent (can run in git worktree)

Subagent model priority: CLAUDE_CODE_SUBAGENT_MODEL env > tool-specified > agent config > inherit parent

Claude Code vs Cursor vs Cline

Full documents: vs Cursor | vs Cline | Feature Matrix

Telemetry — What Data Is Collected

Full document | Source: src/services/analytics/

Dual-Channel Architecture

Claude Code → First-Party (api.anthropic.com/api/event_logging/batch)  [640+ event types]
           → Datadog (us5.datadoghq.com)                              [64 allowed events]
           → BigQuery Metrics (api.anthropic.com/api/claude_code/metrics)

What's Collected

• Environment: platform, arch, terminal, package managers, runtimes, CI detection
• User ID: device_id (de-identified), user_bucket (SHA256 % 30), session_id
• Events: session lifecycle, API calls (model, tokens, latency), tool usage, permission decisions, voice toggles
• NOT collected by default: user prompts (requires OTEL_LOG_USER_PROMPTS=1), full file paths (extension only)

How to Disable

DISABLE_TELEMETRY=1                              # Disable analytics
CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC=1       # Disable all non-API network

Remote Control & Killswitches

Full document | Source: src/services/remoteManagedSettings/

Polling Mechanism

• Endpoint: {BASE_URL}/api/claude_code/settings
• Interval: Every 1 hour
• Cache: ETag-based with SHA256 checksum verification
• Failure mode: Fail-open (continues working with cached/default settings)

Known Killswitches

Remotely Modifiable Behaviors

Data reporting, auto mode, permission bypass, polling frequency, event sampling, model routing, memory system, cache strategy, Bash classification, version limits.

Unreleased Features — Future Roadmap

Full document

Assessed by verifiable code metrics, not subjective percentages:

108 modules are dead-code-eliminated from the npm build and exist only in Anthropic's internal monorepo.

CLAUDE.md Best Practices

Full document | Source: src/context.ts

Loading Order

~/.claude/CLAUDE.md (global) + project/CLAUDE.md (project) + .claude/CLAUDE.md
  → Merged → Injected into system prompt dynamic section → Sent with EVERY request

Recommended Template

# CLAUDE.md

## Tech Stack
- Language: TypeScript 5.x, strict mode
- Framework: Next.js 14, App Router
- Testing: vitest
- Package manager: pnpm

## Conventions
- Components: PascalCase, hooks: camelCase
- Commits: conventional commits (feat/fix/chore)

## Project Structure
- src/app/ — routes
- src/components/ — shared components
- src/lib/ — utilities

What NOT to Write

• Don't repeat system prompt rules (already hardcoded)
• Don't write long descriptions (wastes tokens every request)
• Don't paste API docs (let Claude read the actual files)
• Don't put TODO lists (use conversation, not CLAUDE.md)

Accuracy & Methodology

Based on static source code reading of Claude Code v2.1.88:

• All numeric constants verified against source with file paths and line numbers
• Architecture diagrams are simplified — actual flow may have uncaptured edge cases
• Feature assessments use verifiable metrics (file count, LOC, cross-references)
• Cursor/Cline comparison uses public documentation; internals may differ
• No code was executed or tested at runtime

Found an error? Open an issue with the source file and line number.

Disclaimer

• NOT affiliated with, endorsed by, or maintained by Anthropic
• All original source code is Anthropic's intellectual property
• For educational and security research purposes only
• No proprietary source code is redistributed — all content is original analysis

License

Analysis and documentation: MIT | Original Claude Code source: Anthropic (all rights reserved)