Minovative Mind CLI

The CLI agent that makes Flash models perform like a Pro reasoning model.

Most AI coding tools throw your entire codebase at the biggest model available and hope for the best.

This CLI does the opposite — it uses a custom built agentic system called, Precision-Context Verification (PCV) engine to feed lightweight Flash models exactly the right context via standardized 2–6 concurrent batch tool calling, whole-file and broad-chunk context gathering, single-flight dependency graph deduplication, 3-tier IR/graph context ranking, AST-level targeted reads, and asymmetric outline injection, execute code with pre-implementation blueprints and sandboxed Test-Driven Generation (TDG), verify compilation using polyglot high-signal diagnostic frames, execute Property-Based Testing (PBT) suites, diff-scoped mutation audits, and finite state machine static analysis (PERF-013), and self-correct (if it even needs to), until the build/performance metrics are green.

The result: Genuine Pro reasoning accuracy at Flash-level speed and efficiency.

• Official Website: Main Website
• Pricing & BYOK: Pricing
• Latest Updates: Updates

Quick Start

Prerequisite: Ensure you have Node.js installed on your machine (v18.0.0 or higher).

npm install -g minovative-mind-cli

Update to latest version

npm install -g minovative-mind-cli

1. Subscribe to BYOK ($3.99/mo)

Minovative Mind CLI operates on a transparent Bring Your Own Key (BYOK) subscription for $3.99/month with 0% token markup.

• Subscribe at minovativemind.dev/pricing
• 🎉 Launch Special: Use promo code MMCLI at checkout for 1 Month 100% Free (limited to the first 1,000 developers).

2. Get your Google AI Studio API Key

• Generate your free or paid API key at Google AI Studio.

3. Login & Start Coding

minovative-mind-cli login   # One-click GitHub or Google sign-in with your subscribed account
cd your-project
minovative-mind-cli chat    # Start coding (set your key when prompted or via /config-key)

Works in any terminal — SSH, Docker, CI pipelines, Vim, Windows Command Prompt, PowerShell, anywhere Node runs.

See it in action

• AI builds a Python Generative Art Engine in one prompt
• AI builds a Quiz Game from scratch

🏆 Benchmark Tests

All-Your-Base (C++ Track)

• Difficulty: Hard | Time: 3.5 minutes
• Result: 🟢 PASSED (17/17 assertions)
  • Successfully deduced the O(N+M) conversion algorithm.
  • Properly handled complex edge cases for zero-equivalents (returning an empty vector).
  • Encountered a compilation error due to missing C++ standard library paths on macOS.
  • Autonomously self-corrected by querying the environment, locating the sysroot, and updating clang++ flags.
  • Flawlessly passed the test suite, handling the problem and edge cases on the first try.
• All-Your-Base Summary

Alphametics (Go Track)

• Difficulty: Extreme | Time: 5.2 minutes
• Result: 🟢 PASSED
  • Solved one of the hardest logic puzzles on the platform by engineering a mathematically rigorous pruning algorithm based on the Rearrangement Inequality.
  • Completely bypassed naive backtracking (10! permutations).
  • Achieved sub-millisecond execution speeds (~140 microseconds) with zero heap allocations during recursion.
  • Autonomously navigated the Go toolchain (go test -bench=.).
  • Parsed its own architectural performance warnings to optimize the solution.
• Alphametics Summary

🔬 Automated SWE-bench Evaluation Runner

Evaluate the agent on SWE-bench Lite datasets with automated repo isolation, patch extraction, AST-scoped context minimization, and public-ready telemetry reporting:

# Run SWE-bench evaluation and generate JSON report
minovative-mind-cli eval -i test_instances.jsonl -o predictions.jsonl -r evaluation_report.json --autoClone

# Run with specific repository filter and concurrency
minovative-mind-cli eval -i instances.jsonl --repo astropy/astropy --concurrency 2

Token Economics & Efficiency: With standardized 2–6 concurrent batch tool calling, whole-file/broad-chunk context gathering, single-flight dependency graph deduplication (inFlightBuildPromises), deterministic 3-tier IR context ranking, AST-level symbol chunking (targetElements), asymmetric outline injection, SHA-256 context caching, and in-place historical tool response pruning, benchmark evaluations achieve up to 85%–95% lower token usage per instance, eliminating token-per-minute (TPM) rate limit thrashing while preserving deep reasoning fidelity.

Models

Minovative Mind CLI features native support for Google's Gemini 3.x reasoning capabilities (thinking_config), allocating dynamic thinking budgets before tool execution and code generation. Hot-swap models and customize reasoning levels during a session using /models:

| Model | Default Thinking Level | Best for | | ------------------------- | ---------------------- | ------------------------------------------------------------------------------------------------------------ | | Auto (default) | Dynamic (MEDIUM / HIGH) | Auto (Dynamic Flash Routing: chat on Flash-Lite, execution & investigation on 3.8 Flash with adaptive reasoning) | | Gemini 3.8 Flash | Medium (MEDIUM) | Cutting-edge reasoning, speed & low-latency execution | | Gemini 3.7 Flash | Medium (MEDIUM) | Next-gen performance, reasoning & fast execution | | Gemini 3.6 Flash | Medium (MEDIUM) | Everyday coding — fast and accurate | | Gemini 3.1 Pro | High (HIGH) | Complex architectural changes & deep reasoning | | Gemini 3.5 Flash Lite | Low (LOW) | Best for raw speed, lightweight chat & classification (code execution automatically uses Flash) |

Native Thinking & Reasoning Levels

Gemini 3.x models support native provider-level reasoning control via thinking_config. When switching models or invoking /models, you can interactively customize the thinking depth across three streamlined developer tiers:

• Low (LOW): Fast, lightweight reasoning for routine tasks, minor bug fixes, and quick code edits.
• Medium (MEDIUM): Balanced reasoning depth for multi-file refactoring, debugging, and feature development (default for 3.8 Flash, 3.7 Flash, 3.6 Flash, and Auto).
• High (HIGH): Exhaustive chain-of-thought deliberation for complex architectural transformations, intricate algorithmic puzzles, and large-scale migrations (default for 3.1 Pro).

(Note: Internal Minimal reasoning is strictly reserved for lightweight background micro-agents and is excluded from user-facing selection to ensure optimal coding performance.)

🔑 BYOK (Bring Your Own Key) All-Access

Minovative Mind CLI operates exclusively on a Bring Your Own Key (BYOK) model ($3.99/month flat developer subscription with zero token markups):

• Wholesale AI Pricing: You pay Google AI Studio directly at pure wholesale rates (or $0 on their generous free tier) with no middleman markup or expiring credit packs.
• Supported Provider: Only API keys from Google AI Studio are supported (all Gemini 3.x Flash, Pro, Lite, Thinking, and Experimental models). Vertex AI, OpenAI, and Anthropic keys are not compatible.
• Secure Keychain Storage: Your API key is encrypted and stored locally in your operating system keychain (keytar / local secure storage) and is never sent to Minovative Mind servers.
• Configuration: Set or update your key anytime in the chat session via /config-key.
• Error Handling: If your key is invalid, expired, or you hit rate limits, the CLI will report a BYOK AI Error. Check your key and quota status in the Google AI Studio dashboard.

🧠 Adaptive Persona & Pair Programming Memory Bank

Minovative Mind CLI continuously adapts to your unique developer personality and working habits without adding latency or telemetry bloat:

• Conversational Wavelength & Pairing Dynamics: Learns your relationship model (collaborative peer vs. direct operator), banter and humor affinity, speech formality, apology tolerance, and conversational quirks (e.g. catchphrases like "lgtm" and "ship it").
• Cognitive & Decision Spectrums: Adapts to your architectural preferences (top-down vs. bottom-up), decision autonomy, risk tolerance, and debugging style.
• Zero-Latency Flash-Lite Profiling: A dedicated background gemini-3.5-flash-lite agent classifies turns and reconciles behavioral drift asynchronously.
• 100% Client-Side Privacy: Stored globally at ~/.minovativemind/user_profile.json (0600 permissions), never synced to cloud databases, and fully manageable via /profile.

Session Commands

| Command | What it does | | --------------- | ------------------------------------------------------------------------------------------------------------------------------------------------ | | /config-key | Configure custom Google AI Studio API key (BYOK mode) | | /profile | View, inspect, delete, or reset global adaptive persona memory, pairing dynamics, and AI side-notes | | /models | Hot-swap the active model or configure its native reasoning thinking level (Low, Medium, High) | | /paste | Multi-line input mode (cancel with Ctrl+C) | | /clear | Clear conversation history and reset session settings to safe defaults | | /debug | Toggle internal agent telemetry and diagnostic logging (saved per session) | | /auto-approve | Toggle skipping confirmation prompts for commands (saved per session, automatically resets to 'ask' on new sessions) | | /sub-agents | Toggle the MMAAK Engine for parallel investigation and execution (enabled by default, saved per session, concurrency limited to 3) | | /stats | View session and all-time statistics, token telemetry, active model settings, and live thinking level runtime telemetry | | /commit | Generate a conventional commit message from your diff | | /revert | Undo changes from the last turn, or toggle the revert logger | | /chats | View, resume, rename, or delete previous chat sessions (with session settings restoration, Git branch, auto-approve status, and sub-agent state) | | /workspaces | Manage active workspaces and linked cross-repo aliases | | stop | Abort generation immediately |

Interactive Autocomplete & Keyboard Navigation

When typing @, the interactive terminal autocomplete popup immediately renders:

• 50-Option Pagination: Displays up to 50 ranked suggestions per page with dynamic page indexing (1/50).
• Cyan Navigation Helper: Highlights the action footer in light blue (Tab to insert • ↑/↓ navigate • Esc dismiss) for clear guidance.
• Keyboard Controls:
  • ↑ / ↓ (Up / Down Arrows): Cycle smoothly through ranked suggestions with automatic window scrolling.
  • Tab or Enter: Select and insert the highlighted mention directly at the cursor.
  • Esc: Dismiss the autocomplete dropdown.
  • Backspace / Alphanumeric keys: Filter suggestions in real-time with fuzzy matching.

Supported Mention Types

| Mention Syntax | Category Badge | Type | What it Resolves & Injects | | :--------------------------------------- | :------------------ | :----------------------- | :----------------------------------------------------------------------------------------------------------------------------------------------------------------- | | @<path> or @file:<path> | [file] | File | Injects the complete content of a workspace file (e.g. @src/index.ts or @file:package.json). | | @<path>:<start>-<end> | [lines] | Line Range Slicing | Surgically slices and injects only the specified line numbers (e.g. @src/utils/config.ts:10-50), saving tokens. | | @symbol:<name> or @#<name> | [sym] | AST Symbol | Injects exact definitions of functions, classes, interfaces, types, methods, or structs (e.g. @symbol:getUser or @#AuthService) using polyglot AST extraction. | | @symbol:<path>:<name> | [sym] | Scoped Symbol | Injects a specific symbol definition directly from a designated source file (e.g. @symbol:src/services/auth.ts:validateToken). | | @git:diff or @diff | [git] | Git Diff | Injects the active unstaged working tree diff. | | @git:staged | [git] | Git Staged | Injects currently staged git changes (git diff --staged). | | @git:status | [git] | Git Status | Injects the current branch state and working tree status (git status --short). | | @git:branch | [git] | Git Branch | Injects the current active Git branch and upstream tracking info. | | @git:log | [git] | Git Commit Log | Injects the recent commit history (last 10 commits with hashes, authors, and messages). | | @diagnostics / @problems / @errors | [diag] / [prob] | Diagnostics | Injects active compiler, linter, or syntax error diagnostics. | | @terminal / @console | [term] | Terminal Context | Injects recent terminal environment context, shell environment, and command execution state. | | @<alias>/<path> | [ws] | Cross-Workspace File | Injects file contents from registered external workspaces (e.g. @backend/src/routes.ts or @frontend/src/App.tsx). | | @<alias> | [ws] | Workspace Summary | Injects configuration, manifest overview, and path details for a registered external workspace. |

🌐 Multi-Workspace, Sub-Path Focusing & Security Guardrails

Minovative Mind CLI doesn't restrict you to a single repository. You can logically group multiple external repositories into Master Workspaces (Profiles) containing dedicated Sub-Workspaces (mapped to short aliases like @backend or @frontend).

🌐 Supported Languages

Minovative Mind CLI fundamentally supports ALL programming languages for chat, code generation, and execution, as it relies on Gemini's vast training data.

However, the PCV engine features deep, context-aware analysis across 12 major programming language families. Our core advanced engines—Smart Dependency Tracing, AST Symbol Extraction & Chunking, Property-Based & Mutation Verification, Performance Auditing (across 9 language families), and Ephemeral Analysis Scripts—provide tailored support depending on the language's syntax and runtime model:

| Language Family | Supported Extensions | Dependency Tracing | AST Extraction | Performance Auditing | Ephemeral Analysis | | :--------------------------- | :------------------------------------------- | :----------------: | :------------: | :------------------: | :----------------: | | JavaScript / TypeScript | .js, .jsx, .ts, .tsx, .mjs, .cjs | ✅ | ✅ | ✅ | ✅ | | Python | .py, .pyw | ✅ | ✅ | ✅ | ✅ | | Rust | .rs | ✅ | ✅ | ✅ | ✅ | | Go | .go | ✅ | ✅ | ✅ | ✅ | | C / C++ | .c, .h, .cpp, .hpp, .cc, .cxx | ✅ | ✅ | ✅ | ⚠️ * | | C# | .cs | ✅ | ✅ | ✅ | ❌ | | Java / Kotlin | .java, .kt, .kts | ✅ | ✅ | ✅ | ⚠️ * | | PHP | .php | ✅ | ✅ | ✅ | ❌ | | Ruby | .rb | ✅ | ✅ | ✅ | ❌ | | CSS / SCSS / SASS / LESS | .css, .scss, .sass, .less | ✅ | ❌ | ❌ | ❌ | | Swift | .swift | ✅ | ✅ | ❌ | ❌ | | Dart | .dart | ✅ | ✅ | ❌ | ❌ |

⚠️*Support is limited or requires custom local system tooling/environment setup from the developer's end.

Why do some features only support specific languages?

1. Dependency Tracing & AST Symbol Extraction: Fully supported across major language families. It uses lightning-fast static regex pattern matching and AST outline parsers to resolve local import structures, extract targeted symbol definitions (targetElements), and determine the "blast radius" of code changes without requiring heavy compilations.
2. Performance Auditing: Executed natively across 9 major language families (JavaScript/TypeScript, Python, Go, Rust, PHP, C#, Java, C/C++, and Ruby). It uses zero-dependency, ultra-fast (<50ms) language-aware regex heuristics with comment/string stripping and line-preserving offset tracking to detect severe runtime anti-patterns (such as unbounded loops, synchronous I/O, chained array allocations, unnecessary allocations, unclosed resources, and boolean state flag sprawl via PERF-013).
3. Ephemeral Analysis & Property-Based Probing: Defaults to Node.js as a safe baseline, but natively leverages host-installed runtimes (Python, Rust, Go compilers) and standard libraries whenever available in the workspace. Features direct binary subprocess spawning without shell overhead, zero-disk stdin streaming for interpreted scripts, per-workspace LRU environment caching, 1:1 stack trace line preservation, and dedicated FD-3 out-of-band telemetry streaming for high-precision diagnostic and property-based verification tools (run_debug_script, run_fuzz_probe, check_heap_delta, check_behavioral_drift).

💡 Performance & IDE Optimization

During heavy, multi-turn AI agent sessions, Minovative Mind CLI includes built-in architectural optimizations to eliminate IDE lag, CPU spikes, and dev-server interference:

• Ephemeral Scratch Isolation (os.tmpdir()) & Zero-Disk Stdin Streaming: Diagnostic probes, validation scripts, and benchmark tests execute via direct subprocess spawning with zero-disk stdin streaming (for Node.js, Python, Bash, Ruby, and PHP) or within a sandboxed OS temp directory (os.tmpdir()) rather than writing disposable files into your workspace root. This prevents file watcher churn (fsevents, inotify), avoids Language Server Protocol (LSP) re-indexing storms (TSServer, Pyright, rust-analyzer), and prevents running dev servers (Vite, Turbopack, Nodemon) from triggering unneeded full-page reloads.
• Per-Workspace LRU Environment Caching & Out-of-Band FD-3 Telemetry: The sandboxed execution engine caches runner binaries, virtual environments, and module types in an LRU cache (30s TTL) to minimize process startup latency, while streaming structured emitResult telemetry over a dedicated FD 3 pipe (stdio: ['pipe', 'pipe', 'pipe', 'pipe']) without polluting console standard output or altering 1:1 stack trace line numbers.
• Automatic IDE Watcher Exclusions (.vscode/settings.json): On session startup, the CLI automatically and non-destructively ensures that .minovativemind/**, .tmp/**, and scratch/** are added to files.watcherExclude and search.exclude. This stops IDE background processes from burning CPU on internal telemetry, chat session logs, and cache files.
• Clean Git Status Synchronization: The CLI automatically ensures .minovativemind/, .tmp/, and scratch/ are excluded from .gitignore, .dockerignore, and .npmignore, keeping your IDE's Source Control pane fast and responsive.
• Terminal Render Efficiency: For long-running evaluation sweeps (eval) or massive test outputs, minimizing the terminal pane or running it in the background pauses Electron/xterm.js GPU canvas repaints and frees up system resources.

For full tuning recommendations and configuration details, see the Performance & IDE Optimization Guide.

Legal

Terms of Service · Privacy Policy