program-directory-and-file-definitions

LM Studio Server Information (Default Mode Only)

LM Studio and its pre-loaded models can be reached at "http://127.0.0.1:1234" (overridable via MOTO_LM_STUDIO_BASE_URL). NOTE: The system works without LM Studio. If LM Studio is unavailable, users can configure OpenRouter for all roles. In generic mode (generic_mode=True), workflow inference paths bypass LM Studio — embeddings use FastEmbed and configured roles normalize to OpenRouter, while hidden legacy diagnostics should not be used by hosted UI.

Runtime Root Note

backend/data/ and backend/logs/ are the default desktop roots, not immutable global paths.

• The active backend instance may override mutable roots with MOTO_DATA_ROOT and MOTO_LOG_ROOT
• Exactly one backend process may own a data root at a time; an OS-backed lifespan lease is authoritative even for direct launches and explicit overrides. Chroma is derived cache state: Windows desktop replaces prior-process native state before first open, while hosted Linux retains persistence.
• Runtime root changes after import must use the atomic bind_runtime_roots() contract, which re-derives the complete mutable path graph; long-lived stores resolve against the active root and RAG opens Chroma lazily.
• Secret persistence may be isolated per instance with MOTO_SECRET_NAMESPACE
• Frontend browser persistence may be isolated on shared origins with VITE_MOTO_STORAGE_PREFIX; launch/control-plane config may supply MOTO_FRONTEND_STORAGE_PREFIX and project it into the frontend env
• Hosted protected request size may be capped with MOTO_GENERIC_MAX_REQUEST_BYTES / GENERIC_MAX_REQUEST_BYTES (default 16 MiB)
• Additional local launches may create generated instance folders such as .moto_instances/{instance_id}/data and .moto_instances/{instance_id}/logs

Current High-Level Project Directory Structure and File Descriptions

This is a maintained source-map for important directories and files, not a strict exhaustive inventory of every generated artifact or test helper. project-root/ ├── .github/ # GitHub community health files │ ├── ISSUE_TEMPLATE/ # Public issue forms + security contact routing │ ├── codeql/ # CodeQL analysis configuration │ ├── workflows/ # GitHub Actions workflows, including CodeQL │ └── pull_request_template.md # Default pull request template ├── .moto_instances/ # Generated local multi-instance runtime roots (not source-controlled) │ └── {instance_id}/ │ ├── data/ # Instance-specific mutable data root │ └── logs/ # Instance-specific mutable log root ├── backend/ │ ├── shared/ # SHARED RESOURCES │ │ ├── init.py # Package initialization │ │ ├── config.py # RAGConfig, SystemConfig (instance runtime contract, ports, data/log roots, secret namespace, generic_mode) │ │ ├── models.py # Pydantic models (includes ModelConfig with per-role supercharge_enabled, BoostConfig, WorkflowTask) │ │ ├── lm_studio_client.py # LM Studio HTTP API client + same-base numeric :# instance sharing │ │ ├── openrouter_client.py # OpenRouter HTTP API client (credit exhaustion detection + model/endpoint metadata) │ │ ├── openai_codex_client.py # Desktop-only OpenAI Codex/ChatGPT OAuth client, Codex backend adapter, and Codex model metadata normalizer │ │ ├── xai_grok_client.py # Desktop-only xAI Grok/SuperGrok OAuth client, model metadata normalizer, and typed hard spending/subscription-limit detection │ │ ├── sakana_fugu_client.py # Desktop-only Sakana client (Responses/chat fallback, model metadata, typed quota/entitlement classification) │ │ ├── api_client_manager.py # Unified API router (exact configured-role enforcement, Supercharge, OpenRouter/OAuth/LM Studio fallback, boost) │ │ ├── provider_errors.py # Typed safe provider-route, context-limit, and operator-repair-required errors │ │ ├── boost_manager.py # Singleton boost manager (tracks boost modes: next-count, always-prefer, category; aliases absorbed LeanOJ path-decision tasks into Final Solver boost category) │ │ ├── boost_logger.py # Boost API call logger (persists redacted/default-safe entries to boost_api_log.txt) │ │ ├── workflow_predictor.py # Legacy/shared workflow prediction helper; active coordinators maintain their own workflow_tasks │ │ ├── workflow_start_guard.py # Process-wide async guard and committed owner for atomic mutually-exclusive workflow starts │ │ ├── runtime_root_lock.py # Cross-process OS-backed lease enforcing one backend owner per active data root │ │ ├── sleep_inhibitor.py # Owner-based Windows system/execution Power Requests for active top-level workflows and independently queued manual proof checks │ │ ├── free_model_manager.py # Free model rotation/cooldown singleton (looping + auto-selector backup) │ │ ├── model_error_utils.py # Shared helpers for non-retryable provider/config failures; only recoverable credit exhaustion should pause workflows │ │ ├── provider_pause.py # Process-local provider-credit pause/resume signal used by proof workflows and OpenRouter reset │ │ ├── rag_lock.py # Global RAG operation lock (prevents Aggregator/Compiler collision) │ │ ├── token_tracker.py # Cumulative input/output token tracker singleton with per-model breakdown and research timer │ │ ├── wolfram_alpha_client.py # Wolfram Alpha API client (logs redacted metadata, not raw query/result text) │ │ ├── utils.py # Common utilities │ │ ├── json_parser.py # JSON parsing with sanitization for LLM quirks │ │ ├── response_extraction.py # Central OpenAI-compatible assistant text extraction with content-first reasoning/thinking fallback policy │ │ ├── critique_memory.py # Paper critique persistence (saves up to 10 validator critiques per paper) │ │ ├── critique_prompts.py # Default critique prompt and builder function for validator critiques │ │ ├── secret_store.py # Secure credential persistence via OS keyring (OpenRouter, desktop OAuth tokens, Wolfram Alpha); bypassed in generic mode │ │ ├── runtime_settings.py # Non-secret runtime setting persistence under the active data root (free-model knobs, desktop/default proof knobs, and connectivity toggles) │ │ ├── provider_notification_store.py # Bounded non-secret provider notifications with workflow ownership and stable recovery keys │ │ ├── build_info.py # Build identity resolver (manifest + git HEAD/ZIP stamp + env overrides) │ │ ├── path_safety.py # Safe path resolution helpers (realpath/normpath containment checks) │ │ ├── syntheticlib4_client.py # Contract-first SyntheticLib4 local-snapshot client for release/status/retrieve/account-proof work and API-key scaffolding; inactive runtime has zero records, while explicit fixtures remain test-only │ │ ├── proof_identity.py # moto-proof-identity-v1: canonical theorem whitespace + Lean newline/outer-whitespace hashes, separate from publisher integrity hashes │ │ ├── proof_search/ # Typed records/normalizers; SQLite schema v2 atomic index; composite exact neighborhoods; freshness/hydration/tool service; policy-versioned at-most-64→21→7 run-safe Assistant selection │ │ ├── solution_path/ # Typed run-scoped Progressive Solution Path state, durable serial proposal review engine, and prompt integration helpers │ │ ├── fastembed_provider.py # FastEmbed embedding wrapper (generic mode only, lazy-imported) │ │ ├── lean4_client.py # Lean 4 proof checker client (subprocess gated on lean4_enabled, optional LSP persistent mode gated on lean4_lsp_enabled; offloads temp/workspace filesystem operations from the FastAPI event loop) │ │ ├── lean_proof_integrity.py # Shared post-Lean integrity gate (rejects fake axiom/constant/opaque devices and validates theorem-statement alignment) │ │ ├── brainstorm_proof_gate.py # Shared Lean 4 gate for optional brainstorm proof candidates before normal validation; rejects missing novelty/rationale metadata before Lean cost and preserves Lean-accepted real artifacts for novelty/validator review │ │ └── smt_client.py # Z3/SMT launcher-managed subprocess wrapper (gated on smt_enabled; never authoritative on its own) │ ├── aggregator/ # AGGREGATOR │ │ ├── init.py │ │ ├── core/ │ │ │ ├── init.py │ │ │ ├── rag_manager.py # 4-stage RAG pipeline orchestrator (ChromaDB calls and heavy retrieval scoring are offloaded/snapshot-based to keep GUI/API routes responsive) │ │ │ ├── coordinator.py # Manages 1-10 submitters + 1 validator (default 3, configurable per-submitter) │ │ │ ├── queue_manager.py # Submission queue FIFO/batch/count/clear helper; coordinator owns submitter pause decisions │ │ │ └── context_allocator.py # Direct injection vs RAG routing (tries direct first, offloads to RAG only when doesn't fit). Includes allocate_cleanup_review_context() which NEVER skips due to size - uses RAG when database too large. │ │ ├── ingestion/ │ │ │ ├── init.py │ │ │ ├── chunker.py # Multi-config chunking (256/512/768/1024) │ │ │ ├── pipeline.py # Document ingestion pipeline │ │ │ ├── normalizer.py # Text normalization │ │ │ └── metadata_extractor.py # Extract metadata from chunks │ │ ├── validation/ │ │ │ ├── init.py │ │ │ ├── contradiction_checker.py # Detect contradictions │ │ │ └── json_validator.py # Validate JSON responses │ │ ├── memory/ │ │ │ ├── init.py │ │ │ ├── shared_training.py # Validator-distributed database (accepted submissions) │ │ │ ├── local_training.py # Per-submitter rejection logs (last 5) │ │ │ └── event_log.py # Persistent event log (acceptances, rejections, cleanup removals) │ │ ├── agents/ │ │ │ ├── init.py │ │ │ ├── submitter.py # Submitter agent (parallel, cyclic chunk sizes) │ │ │ └── validator.py # Validator agent (sequential validation) │ │ └── prompts/ │ │ ├── init.py │ │ ├── submitter_prompts.py # Submitter system prompts + JSON schemas │ │ └── validator_prompts.py # Validator system prompts + JSON schemas │ │ │ ├── compiler/ # COMPILER (Phase 2) │ │ ├── init.py # Package initialization │ │ ├── core/ │ │ │ ├── init.py # Package initialization │ │ │ ├── compiler_coordinator.py # Orchestrates sequential Markov chain workflow │ │ │ └── compiler_rag_manager.py # Compiler-specific RAG wrapper (user-configurable context per role) │ │ ├── agents/ │ │ │ ├── init.py # Package initialization │ │ │ ├── writer_submitter.py # 3 modes: construction, outline, review │ │ │ └── high_param_submitter.py # Rigor & Proofs role (rigor/proof/critique generation) │ │ ├── validation/ │ │ │ ├── init.py # Package initialization │ │ │ └── compiler_validator.py # Validates coherence, rigor, placement (plus rigor_lean_placement mode for Lean-4 theorem placement) │ │ ├── prompts/ │ │ │ ├── init.py # Package initialization │ │ │ ├── outline_prompts.py # Outline generation & update prompts │ │ │ ├── construction_prompts.py # Paper construction prompts │ │ │ ├── review_prompts.py # Paper review/cleanup prompts │ │ │ └── rigor_prompts.py # Rigor theorem-discovery + placement prompts (Lean-4-verified flow) │ │ └── memory/ │ │ ├── init.py # Package initialization │ │ ├── outline_memory.py # Current outline state (direct inject/RAG) │ │ ├── paper_memory.py # Current paper state (direct inject/RAG). Skeleton markers: ABSTRACT_PLACEHOLDER, INTRO_PLACEHOLDER, CONCLUSION_PLACEHOLDER, THEOREMS_APPENDIX_START/END (wraps Lean-4-verified theorem entries), PAPER_ANCHOR. │ │ ├── critique_memory.py # Accepted critiques database for peer review phase │ │ ├── critique_rejection_memory.py # Last 5 critique rejection feedback logs │ │ └── compiler_rejection_log.py # Last 10 rejections & acceptances │ │ │ ├── autonomous/ # AUTONOMOUS RESEARCH (Phase 3) │ │ ├── init.py # Package initialization │ │ ├── proof_pruning_evidence.py # Pure deterministic semantic-pruning evidence ranking, projection, token estimation, and evidence/dependency/descriptor fingerprints │ │ ├── core/ │ │ │ ├── init.py # Package initialization │ │ │ ├── autonomous_coordinator.py # Orchestrates the Tier 1 → Tier 2 → Tier 3 autonomous workflow (invokes proof verification checkpoints after brainstorm/Tier 2 paper completion when lean4_enabled) │ │ │ ├── autonomous_rag_manager.py # Autonomous-specific RAG wrapper │ │ │ ├── proof_round_driver.py # Shared one-round, bounded automatic multi-round, and unbounded continuous-pruning policies; candidate and Lean execution stay in ProofVerificationStage │ │ │ ├── proof_verification_stage.py # Orchestrates impact-first identification → Lean attempts (3 full + 2 tactic) → integrity/downshift → post-Lean novelty/storage; optional SMT/Mathlib; per-source lock │ │ │ ├── proof_run_manager.py # Owns one-round and continuous manual proof-run lifecycle/status, Stop, terminal cleanup, source reservations, and sleep ownership │ │ │ ├── proof_pruning_coordinator.py # Run-scoped bounded semantic review scheduling, generation-safe coalescing, durable migration, stale-fenced commit, and lifecycle drain │ │ │ ├── proof_novelty.py # Shared proof novelty assessment helper used by autonomous proof verification and compiler rigor │ │ │ ├── proof_registration.py # Shared registration helper for verified Lean proofs from autonomous, compiler, aggregator, and LeanOJ flows │ │ │ └── proof_dependency_extractor.py # Parses verified Lean 4 code to extract ProofDependency records (imports, Mathlib lemmas, MOTO-origin refs) │ │ ├── agents/ │ │ │ ├── init.py # Package initialization │ │ │ ├── topic_selector.py # Topic selection submitter (new/continue) │ │ │ ├── topic_validator.py # Topic selection validator │ │ │ ├── completion_reviewer.py # Brainstorm completion review (SPECIAL SELF-VALIDATION) │ │ │ ├── reference_selector.py # Reference paper selection workflow │ │ │ ├── paper_title_selector.py # Paper title selection │ │ │ ├── proof_identification_agent.py # Extracts impact-first theorem candidates with expected novelty/prompt-relevance/anti-known-result rationale; skips not_novel or missing-tier candidates before Lean cost │ │ │ ├── proof_candidate_list_validator.py # Independent identity-complete pre-Lean Validator for whole candidate lists and the 75% novelty gate │ │ │ ├── proof_formalization_agent.py # Generates Lean 4 proof scripts for candidates with mandatory full source context plus novelty metadata, Mathlib hints, and SMT hints when enabled │ │ │ ├── proof_pruning_agent.py # Fail-closed semantic proposer/Validator over bounded candidate-centered or conservatively arbitrated section evidence; returns fenced non-mutating commit intent │ │ │ ├── lemma_search_agent.py # Mathlib lemma search agent (Build 2) — surfaces relevant existing lemmas for formalization prompts │ │ │ └── final_answer/ # TIER 3 - Final Answer Generation Agents │ │ │ ├── init.py # Package initialization │ │ │ ├── certainty_assessor.py # Assesses "known certainties" from Tier 2 papers │ │ │ ├── answer_format_selector.py # Selects short-form vs long-form answer │ │ │ └── volume_organizer.py # Organizes volume structure (long-form) │ │ ├── validation/ │ │ │ ├── init.py # Package initialization │ │ │ └── paper_redundancy_checker.py # Paper library redundancy review │ │ ├── prompts/ │ │ │ ├── init.py # Package initialization │ │ │ ├── topic_prompts.py # Topic selection & validation prompts │ │ │ ├── topic_exploration_prompts.py # Builds aggregator user prompt for topic exploration phase │ │ │ ├── completion_prompts.py # Completion review & self-validation prompts │ │ │ ├── paper_reference_prompts.py # Reference selection prompts │ │ │ ├── paper_title_exploration_prompts.py # Builds aggregator user prompt for paper title exploration phase │ │ │ ├── paper_title_prompts.py # Paper title selection prompts │ │ │ ├── paper_redundancy_prompts.py # Paper redundancy review prompts │ │ │ ├── paper_continuation_prompts.py # Brainstorm multi-paper continuation decision prompts │ │ │ ├── proof_prompts.py # Proof identification/formalization/statement-alignment/novelty prompts + failure-hint direct injection helper │ │ │ ├── proof_pruning_prompts.py # Isolated proposer/Validator/repair contracts for proof live-context pruning │ │ │ └── final_answer_prompts.py # TIER 3 - Final answer assessment/selection/volume prompts │ │ └── memory/ │ │ ├── init.py # Package initialization │ │ ├── brainstorm_memory.py # Per-brainstorm database management (includes retroactive edit/remove/add during paper compilation) │ │ ├── paper_library.py # Paper library management (Tier 2) │ │ ├── research_metadata.py # Research metadata (brainstorms + papers associations, proof runtime config snapshot) │ │ ├── autonomous_rejection_logs.py # Topic selection & completion feedback logs │ │ ├── topic_exploration_memory.py # In-memory candidate DB for topic exploration phase │ │ ├── paper_model_tracker.py # Per-paper model usage tracking and author attribution │ │ ├── autonomous_api_logger.py # Autonomous API call logger singleton │ │ ├── proof_database.py # Session-aware Lean 4 verified-proof storage (records, Lean sources, novelty/reverse-dependency indexes, owning-run live-context state, revisioned pruning snapshots); cross-session library access via list_proof_library() and get_library_proof() │ │ ├── final_answer_memory.py # TIER 3 - Final answer state & volume management │ │ └── session_manager.py # Prompt-based session folder organization │ │ │ ├── leanoj/ # LEANOJ PROOF SOLVER (proof-only top-level mode; no paper writing) │ │ ├── init.py │ │ ├── prompts.py # LeanOJ topic/batched-topic/brainstorm/prune/path/final-solver JSON prompts; brainstorm validation assigns final-context roles; prune review conservatively asks whether outdated/redundant memory should be removed, updated, or supplemented with one compact corrective idea; final solver edits single-route durable master proofs and reviews Lean-accepted final code before stop │ │ └── core/ │ │ ├── init.py │ │ ├── leanoj_context.py # LeanOJ full artifact persistence, direct-first allocation, final-context routing (verified helpers + active plans direct; refuted constructions separated as compact warnings), session-scoped RAG indexing/retrieval, final-cycle packets, and clear/resume cleanup │ │ └── leanoj_coordinator.py # Parallel topic + brainstorm submitters, batch validators, resumable persistence, partial proof capture, context-role classification, single-route durable master_proof.lean editing, watchdog, final Lean verification, and semantic final-solver review │ │ │ ├── scripts/ # Utility scripts │ │ ├── cache_openrouter_models.py # Caches OpenRouter models with mapping display_name -> api_id │ │ └── startup/ # Legacy/internal startup helper scripts kept out of repo root │ │ │ ├── api/ │ │ ├── init.py │ │ ├── main.py # FastAPI app entry point (lifespan reads generic_mode, fail-closes hosted auth env, ensures desktop API token, skips LM Studio test in generic mode) │ │ ├── middleware.py # CORS, error handling, desktop token/origin checks, proxy auth validation, hosted body-size cap + actual-body hash check │ │ ├── proxy_auth.py # Shared generic-mode proxy auth helpers (allowlist + HMAC over method/path/query/verified body hash for REST/WebSocket) │ │ └── routes/ │ │ ├── init.py │ │ ├── aggregator.py # Aggregator API endpoints (includes /events) │ │ ├── compiler.py # Compiler API endpoints │ │ ├── autonomous.py # Autonomous Research API endpoints │ │ ├── leanoj.py # LeanOJ Proof Solver API endpoints (/api/leanoj/*: start with matching-progress resume, stop, status, master-proof draft/edit summaries, current-run proofs, cross-session library, skip-brainstorm, force-brainstorm, clear) │ │ ├── boost.py # Boost API endpoints (enable/disable/toggle/status + OpenRouter provider endpoint metadata) │ │ ├── workflow.py # Workflow predictions plus typed Progressive Solution Path snapshot and generation-fenced repair-resume endpoints │ │ ├── update.py # Update/check endpoints for launcher/updater state (POST /api/update/pull, GET /api/update/pull-status) │ │ ├── download.py # PDF generation endpoint via Playwright (desktop only; sanitize/block external requests; returns 501 in generic mode) │ │ ├── openrouter.py # OpenRouter API endpoints (global key, models/providers via header/body keys only, LM Studio availability, model cache, reset exhaustion) │ │ ├── cloud_access.py # OpenRouter/OAuth support endpoints including desktop OAuth login/model listing │ │ ├── websocket.py # WebSocket for real-time updates (generic proxy auth or desktop one-time tickets before accept) │ │ ├── features.py # GET /api/features — shared build identity plus stable capability flags; GET /api/update-notice — launcher/runtime-refreshed update notice │ │ ├── proofs.py # Proof database + Lean 4/SMT runtime + manual proof-check/run discovery/control + live-context mutation + certificate/dependency routes; current proof lists expose proof_set_revision, and bounded run collection/source lookup support reconnect recovery │ │ ├── proof_search.py # Unified overview/search/reindex, bounded detail hydration (session_id/search_id/run_id), latest Assistant pack, and paginated support-lineage routes; public search caps at 7 │ │ ├── syntheticlib4.py # SyntheticLib4 corpus status/release/refresh/safe local import/reindex/retrieve-batch/account-proof routes plus production OAuth placeholders; public retrieve-batch is capped at 7 records │ │ ├── connectivity.py # Non-secret grouped connectivity status/toggle routes for inference providers and optional skills │ │ └── health.py # GET /api/health — readiness/liveness probe with slim instance/build metadata │ │ │ ├── data/ # Persistent data storage │ │ ├── user_uploads/ # User-uploaded files │ │ ├── model_cache.json # Auto-generated by cache script - Maps display names to OpenRouter API IDs for profile system │ │ ├── rag_shared_training.txt # Accepted submissions (aggregator output) │ │ ├── aggregator_event_log.txt # Persistent event log (key events survive restarts) │ │ ├── aggregator_stats.json # Persistent stats (acceptances, rejections, cleanup stats) │ │ ├── compiler_outline.txt # Current paper outline │ │ ├── compiler_paper.txt # Current paper being constructed │ │ ├── compiler_last_10_rejections.txt # Last 10 compiler rejections │ │ ├── compiler_last_10_acceptances.txt # Last 10 compiler acceptances │ │ ├── Summary_Of_Last_5_Validator_Rejections_For_Submitter_.txt # Aggregator submitter rejections (1-10, dynamically created per submitter) │ │ ├── auto_brainstorms/ # Autonomous Research - Tier 1 (Brainstorm databases) │ │ │ ├── brainstorm_{topic_id}.txt # Per-brainstorm accepted submissions │ │ │ ├── brainstorm_{topic_id}metadata.json # Brainstorm metadata │ │ │ ├── brainstorm{topic_id}submitter_1_rejections.txt # Per-brainstorm submitter rejections │ │ │ ├── brainstorm{topic_id}submitter_2_rejections.txt │ │ │ ├── brainstorm{topic_id}submitter_3_rejections.txt │ │ │ └── completion_feedback{topic_id}.txt # Completion review feedback (last 5) │ │ ├── auto_papers/ # Autonomous Research - Tier 2 (Finished papers) │ │ │ ├── paper_{paper_id}.txt # Full paper content │ │ │ ├── paper_{paper_id}abstract.txt # Abstract only │ │ │ ├── paper{paper_id}source_brainstorm.txt # Cached brainstorm database │ │ │ ├── paper{paper_id}last_10_rejections.txt # Compiler rejections for this paper │ │ │ ├── pruned/ # Pruned papers preserved for user download, excluded from model context │ │ │ │ ├── pruned_paper{paper_id}.txt # Pruned full paper with top-of-file PRUNED PAPER banner │ │ │ │ └── pruned_paper_{paper_id}metadata.json # Pruned metadata/reason │ │ │ └── archive/ # Legacy archived (redundant) papers, treated as pruned history │ │ │ └── paper{paper_id}.txt │ │ ├── auto_final_answer/ # Autonomous Research - Tier 3 (LEGACY - replaced by auto_sessions) │ │ │ ├── final_answer_state.json # Tier 3 state (crash recovery) │ │ │ ├── volume_organization.json # Volume structure (long-form) │ │ │ ├── tier3_rejections.txt # Tier 3 rejection log (last 10) │ │ │ ├── chapter_{index}paper.txt # Gap/intro/conclusion papers │ │ │ ├── chapter{index}outline.txt # Chapter outlines │ │ │ └── final_volume.txt # Assembled volume (long-form) │ │ ├── auto_sessions/ # Autonomous Research - Session-based folder organization │ │ │ └── {sanitized_prompt}{timestamp}/ # Per-session folder │ │ │ ├── brainstorms/ # Tier 1 brainstorm databases │ │ │ ├── papers/ # Tier 2 completed papers plus pruned/ preserved context-excluded papers │ │ │ ├── final_answer/ # Tier 3 final answer data │ │ │ ├── proofs/ # Lean 4 verified-proof records (proofs_index.json, proof_{proof_id}.json, proof_{proof_id}_lean.lean; generated IDs currently include the proof_ prefix) │ │ │ ├── session_metadata.json # Session info (prompt, created_at, status) │ │ │ ├── session_stats.json # Session statistics │ │ │ └── workflow_state.json # Workflow state for crash recovery │ │ ├── proofs/ # Legacy (non-session) Lean 4 proof storage (mirrors per-session proofs/ layout) │ │ ├── manual_proofs/ # Active manual Aggregator/Compiler proof storage; archived/cleared on manual run reset so old proofs do not enter new manual prompts │ │ ├── manual_proof_runs/ # Archived manual proof runs for history/library viewing only, never active prompt context │ │ ├── proof_search/ # Derived schema/identity-versioned SQLite/FTS index, atomically rebuildable from canonical proof sources │ │ ├── syntheticlib4/ # Planned authorized SyntheticLib4 local snapshots/status/cache under the active data root │ │ ├── leanoj_sessions/ # LeanOJ run state (state.json, master_proof.lean, master_proof_edits.jsonl, master_proof_snapshots.jsonl, phase counters, proof fragments, attempts, verified final Lean code; stop/crash resumes unless cleared) │ │ ├── leanoj_partial_proofs/ # LeanOJ partial/supporting proof scaffold JSONL store, keyed by session │ │ ├── leanoj_artifacts/ # LeanOJ full-memory artifact logs (accepted ideas with context_role metadata, verified/partial/failed proof fragments, final attempts, final-cycle packets) used for direct-first RAG allocation │ │ ├── auto_research_metadata.json # Autonomous Research metadata (LEGACY - now in session folders) │ │ ├── auto_research_stats.json # Autonomous Research statistics (LEGACY - now in session folders) │ │ ├── auto_workflow_state.json # Autonomous Research workflow state (LEGACY - now in session folders) │ │ ├── auto_research_topic_rejections.txt # Topic selection rejections (last 5) │ │ └── chroma_db/ # ChromaDB persistent storage │ │ │ └── logs/ # Application logs │ ├── frontend/ │ ├── src/ │ │ ├── components/ │ │ │ ├── aggregator/ # AGGREGATOR │ │ │ │ ├── AggregatorInterface.jsx # User prompt, file upload, start/stop, developer-gated Creativity Emphasis Boost │ │ │ │ ├── AggregatorSettings.jsx # Model selection, context sizes, capability-gated LM/OpenRouter UI │ │ │ │ ├── AggregatorLogs.jsx # Metrics, acceptance rates, queue; loads persisted events on mount │ │ │ │ └── LiveResults.jsx # Real-time accepted submissions view │ │ │ │ │ │ │ ├── compiler/ # COMPILER │ │ │ │ ├── CompilerInterface.jsx # Manual Compiler prompt, allowed-output controls, start/stop, and status │ │ │ │ ├── CompilerSettings.jsx # Compiler role selections (Validator, Writing, Rigor & Proofs, Assistant), capability-gated LM/OpenRouter UI; deprecated critique fields mirror Rigor & Proofs │ │ │ │ ├── CompilerLogs.jsx # Metrics: construction vs rigor, miniscule edits │ │ │ │ └── LivePaper.jsx # Real-time paper viewing, save draft, word count │ │ │ │ │ │ │ ├── autonomous/ # AUTONOMOUS RESEARCH │ │ │ ├── AutonomousResearchInterface.jsx # Main control: research prompt, start/stop, current tier, developer-gated Creativity Emphasis Boost │ │ │ ├── AutonomousResearch.css # Autonomous research styles │ │ │ ├── BrainstormList.jsx # List all brainstorm topics with status │ │ │ ├── PaperLibrary.jsx # Grid view of completed papers (title + abstract) │ │ │ ├── AutonomousResearchSettings.jsx # Autonomous settings: models/providers/profiles, free-model controls, Tier 3 toggle, Wolfram, Lean/SMT proof runtime, developer-only raw/Supercharge UI │ │ │ ├── AutonomousResearchLogs.jsx # Metrics/per-submitter stats, event log, combined API call logs │ │ │ ├── LivePaperProgress.jsx # Real-time Tier 2 paper display (embedded in interface) │ │ │ ├── LiveTier3Progress.jsx # Real-time Tier 3 final answer display (embedded in interface) │ │ │ ├── FinalAnswerView.jsx # TIER 3 - Final answer tab (separate tab for completed answers) │ │ │ ├── FinalAnswerLibrary.jsx # All sessions final answer library viewer │ │ │ ├── FinalAnswerLibrary.css # Final answer library styles │ │ │ ├── ArchiveViewerModal.jsx # Research lineage archive viewer (papers + brainstorms) │ │ │ ├── ArchiveViewerModal.css # Archive viewer styles │ │ │ ├── MathematicalProofs.jsx # Flat live-session per-proof tab reused by Autonomous and Manual modes; shared run controls plus readable owning-run prune status/actions, dependency graph, and scoped exports │ │ │ ├── MathematicalProofs.css # Proof library styles │ │ │ ├── ProofCheckModeModal.jsx # Accessible shared one-round/continuous proof-check chooser used by all six check entry points │ │ │ ├── ProofRunStatusControls.jsx # Shared source/run round, pruning, Stop, and terminal status controls │ │ │ ├── ProofGraph.jsx # Hand-rolled SVG proof dependency graph view │ │ │ ├── ProofGraph.css # Proof graph styles │ │ │ ├── ProofNotificationStack.jsx # Persistent popup notifications for novel proof discoveries │ │ │ ├── ProofLibrary.jsx # Historical proof library with novelty filter, collapsed run/prompt groups, full expanded proof content, and read-only persisted live-context prune provenance │ │ │ ├── ProofLibrary.css # Proof library viewer styles │ │ │ ├── Stage2PaperHistory.jsx # Tier 2 paper history list (grouped per research run; sub-tab inside CompletedWorksLibrary) │ │ │ └── Stage2PaperHistory.css # Tier 2 paper history styles │ │ │ │ │ │ └── leanoj/ # LEANOJ PROOF SOLVER UI │ │ │ ├── LeanOJInterface.jsx # Prompt/template input, start/resume, stop, skip/force brainstorm, clear progress, live status, developer-gated Creativity Emphasis Boost, verified Lean output │ │ │ ├── LeanOJSettings.jsx # LeanOJ-specific model profiles/settings; grouped UI controls map user-facing roles to the underlying LeanOJ request role keys │ │ │ ├── LeanOJBrainstorms.jsx # LeanOJ accepted ideas/recursive brainstorm memory viewer │ │ │ ├── LeanOJLogs.jsx # LeanOJ API-call log wrapper; live event stream and proof fragments are shown in the interface/proof tabs │ │ │ ├── LeanOJMasterProof.jsx # Master proof draft tab (on-demand draft, metadata, edit history, download) │ │ │ ├── LeanOJMasterProof.css # Master proof draft tab styles │ │ │ ├── LeanOJMathematicalProofs.jsx # Flat current-run verified proof/proof-fragment viewer; suppresses run-prompt fallback fields without truncating proof content │ │ │ ├── LeanOJProofLibrary.jsx # Historical completed-run library with collapsed prompt groups and bounded prompt-header previews; excludes the currently loaded run │ │ │ └── index.js # LeanOJ component exports │ │ │ │ │ ├── StartupProviderSetupModal.jsx # Post-disclaimer startup chooser for OpenRouter or LM Studio setup; desktop OAuth is shown only as an after-startup add-on because embeddings require OpenRouter/LM Studio (OpenRouter-only in generic mode) │ │ ├── ConnectivityPanel.jsx # Grouped top-right connectivity launcher for inference providers and optional skills │ │ ├── OpenRouterApiKeyModal.jsx # OpenRouter/OAuth modal for OpenRouter API key and desktop OAuth logins │ │ ├── SyntheticLib4AccessModal.jsx # SyntheticLib4 coming-soon proof-corpus explainer │ │ ├── WolframAlphaAccessModal.jsx # Wolfram Alpha App ID and enable/disable modal │ │ ├── LMStudioConnectivityModal.jsx # LM Studio status/setup modal │ │ ├── AgentConversationMemoryModal.jsx # User-facing stored-proof memory toggle/status modal │ │ ├── PaperCritiqueModal.jsx # Modal for displaying validator paper critiques (ratings, feedback, history) │ │ ├── CritiqueNotificationStack.jsx # Persistent popup notifications for high-scoring critiques (≥6.25 avg) │ │ ├── CreditExhaustionNotificationStack.jsx # Persistent red notifications for OpenRouter credit exhaustion with "Retry OpenRouter" reset button │ │ ├── CodexOAuthNotificationStack.jsx # Persistent desktop Codex OAuth/provider-repair notifications │ │ ├── ModelErrorNotificationStack.jsx # Durable workflow-scoped model/configuration repair notifications │ │ ├── OpenRouterFreeModelsControl.jsx # Shared free-model filter, looping, and auto-selector controls │ │ ├── ApiCallLogs.jsx # Shared combined API-call log viewer │ │ ├── BoostControlModal.jsx # Modal for boost configuration (next-X, category, always-prefer) with mode-aware copy │ │ ├── BoostControlModal.css # Boost control modal styles │ │ ├── WorkflowPanel.jsx # Boost/token/Assistant panel plus activation-gated, stale-fenced Progressive Solution Path card and lifecycle refresh │ │ ├── WorkflowPanel.css # Boost controls panel and Assistant Memory Bank styles │ │ ├── SolutionPathModal.jsx # Advisory route viewer with repair guidance, settings routing, retry, focus restoration, and accessible status │ │ ├── SolutionPathModal.css # Progressive Solution Path modal styles │ │ ├── TextFileUploader.jsx # User file upload component │ │ ├── TextFileUploader.css # File uploader styles │ │ ├── OpenRouterPrivacyWarningModal.jsx # Privacy policy error modal (OpenRouter data sharing, capability-aware alternatives) │ │ ├── HelpTooltip.jsx # Shared portal-based help tooltip component (used across settings/interfaces) │ │ ├── ProofStrengthBadge.jsx # Shared PS badge/tooltip for highlighted proof-strength models and primary proof-creation roles │ │ ├── settings-common.css # Shared settings panel styles │ │ ├── critique-modal.css # Paper critique modal styles │ │ │ │ │ ├── services/ │ │ │ ├── api.js # Backend API calls (includes openRouterAPI, /api/features, proof routes, LeanOJ API, and cross-session proof library helpers) │ │ │ └── websocket.js # WebSocket connection │ │ │ │ │ ├── hooks/ │ │ │ └── useProofCheckRuntime.js # Shared source-keyed proof-run controller: readiness, queueing, discovery/reload reconciliation, lifecycle events, and generation-fenced controls │ │ │ │ │ ├── utils/ │ │ │ ├── downloadHelpers.js # PDF/raw download helpers (Playwright backend PDF) │ │ │ ├── modelCache.js # Frontend model cache utilities (display_name → api_id lookup) │ │ │ ├── openRouterSelection.js # Shared OpenRouter selector auto-fill helpers (context/output from model + host metadata) │ │ │ ├── autonomousProfiles.js # Shared autonomous recommended-profile definitions and persistence helpers │ │ │ ├── leanojProfiles.js # LeanOJ-specific recommended/user profile definitions, persistence helpers, and request builder (topic generation uses all submitters; legacy topic_generator/selector is sourced from Brainstorm Submitter 1; legacy path_decider request field is derived from Final Proof Solver) │ │ │ ├── safeStorage.js # Defensive browser-storage readers for small UI preferences │ │ │ ├── activityStyles.js # Shared live-activity formatting/styling helpers │ │ │ ├── activityPersistence.js # Bounded activity persistence helpers │ │ │ ├── liveActivityPersistence.js # Instance-prefixed durable live-activity storage and recovery │ │ │ ├── workflowLifecycle.js # Workflow lifecycle identity and stale-event reconciliation helpers │ │ │ ├── proofPresentation.js # Live-context prune classification, risk warnings, and revision/identity-fenced mutation payloads │ │ │ ├── manualProofSources.js # Manual proof-source identity and lookup helpers │ │ │ ├── manualLogRouting.js # Manual workflow proof/activity routing helpers │ │ │ ├── solutionPathPresentation.js # Progressive Solution Path UI state and stale-safe presentation helpers │ │ │ ├── credentialProviderRefresh.js # Mounted-settings credential/provider refresh coordination │ │ │ ├── promptDraftStorage.js # Durable prompt-draft storage with large-value offload │ │ │ ├── oauthProviders.js # Shared desktop OAuth/subscription provider registry helpers │ │ │ ├── runtimeConfig.js # Frontend runtime helpers (instance storage prefix, active data-root display, instance ID) │ │ │ ├── researchRunHistory.js # Groups Tier 2 papers + final answers into per-run history entries for Stage2PaperHistory/FinalAnswerLibrary │ │ │ └── disclaimerHelper.js # Frontend-only disclaimer injection for brainstorm/paper views │ │ │ │ │ ├── App.jsx # Main shell, capabilities/mode routing, global solution-path modal/settings ownership, and workflow-scoped path activity dispatch │ │ ├── index.css # Styles │ │ └── index.jsx # React entry point │ │ │ ├── package.json │ └── vite.config.js │ ├── tests/ # Regression/security tests for launcher, update, API hardening, proof/compiler behavior, and orchestration helpers │ └── test_.py # Pytest modules; keep focused on behavior and path/security regressions │ ├── requirements.txt # Python dependencies (default mode) ├── requirements-generic.txt # Additive deps for generic mode (-r requirements.txt + fastembed + onnxruntime) ├── Dockerfile # Canonical hosted generic-mode container contract (python:3.12-slim, API-only backend image) ├── .dockerignore # Hosted image build-context exclusions (desktop/frontend/dev artifacts stay out of the container) ├── docker/ │ └── entrypoint.sh # Hosted backend startup wrapper (defaults hosted envs, creates runtime dirs, delegates to backend.api.main) ├── package.json # Root scripts ├── moto-update-manifest.json # Build 0 updater/build identity manifest committed on main ├── SECURITY.md # Security policy and private vulnerability reporting ├── Click To Launch MOTO.bat # The authoritative Windows launcher entrypoint (bootstraps Python 3.10+ via winget when missing, refreshes/relaunches if needed, then delegates to moto_launcher.py) ├── linux-ubuntu-launcher.sh # Linux/Ubuntu launcher entrypoint (thin bash wrapper that delegates to moto_launcher.py) ├── moto_launcher.py # Internal Python launcher orchestration (update check, runtime resolution, dependency install, service startup) ├── moto_updater.py # Build 1 updater helper (manifest fetch, install classification, ZIP/git apply flow, launcher state tracking) └── .moto_launcher_state.json # Gitignored local launcher state (tracks active service-window PIDs and runtime roots to block unsafe update-apply)