mentat-sr 0.1.0__tar.gz

mentat_sr-0.1.0/.claude/settings.local.json

@@ -0,0 +1,30 @@
+{
+  "permissions": {
+    "allow": [
+      "Bash(wc:*)",
+      "Bash(uv run python:*)",
+      "Bash(echo:*)",
+      "Bash(uv add:*)",
+      "Bash(uv run pytest:*)",
+      "Bash(uv run mentat probe:*)",
+      "Bash(uv run:*)",
+      "Bash(uv sync:*)",
+      "Bash(git commit:*)",
+      "Bash(chmod:*)",
+      "Bash(git -C /opt/nvme/home/shelven/Documents/mentat diff /opt/nvme/home/shelven/Documents/mentat/mentat/probes/json_probe.py)",
+      "Bash(/tmp/test_large_json.json:*)",
+      "Bash(ls:*)",
+      "Bash(python -c:*)",
+      "Bash(python3 -c \":*)",
+      "Bash(git checkout mentat/core/hub.py mentat/storage/vector_db.py)",
+      "Bash(git add mentat/probes/_utils.py mentat/probes/docx_probe.py mentat/probes/json_probe.py mentat/probes/markdown_probe.py mentat/probes/web_probe.py mentat/probes/instruction_templates.py mentat/probes/test_instruction_templates.py mentat/core/hub.py)",
+      "Bash(git add mentat/core/queue.py mentat/core/embeddings.py mentat/librarian/engine.py mentat/core/hub.py mentat/storage/vector_db.py tests/test_async_workflow.py tests/test_performance.py)",
+      "Bash(git add mentat/server.py mentat/core/access_tracker.py mentat/cli.py mentat/__init__.py pyproject.toml uv.lock mentat/storage/cache.py mentat/core/hub.py)",
+      "Bash(git status -u)",
+      "Bash(git add mentat/probes/pdf_probe.py benchmarks/benchmark.py benchmarks/CONCLUSION.md benchmarks/results.json benchmarks/1706.03762v7.pdf)",
+      "Bash(grep -n \"track_access\\\\|_on_access_promote\\\\|summarize_doc\\\\|hot.*promote\" /opt/nvme/home/shelven/Documents/mentat/tests/*.py)",
+      "Bash(git add CLAUDE.md TODO.md mentat/__init__.py mentat/cli.py mentat/core/access_tracker.py mentat/core/hub.py mentat/server.py mentat/skill.py tests/test_access_tracker.py tests/test_smoke.py tests/test_skill.py)",
+      "Bash(git add:*)"
+    ]
+  }
+}

mentat_sr-0.1.0/.env.example

@@ -0,0 +1,54 @@
+# =============================================================================
+# Mentat Configuration
+# Copy this file to .env and fill in the values.
+# =============================================================================
+
+# --- Summary Model (Phase 1 — chunk summarisation) ---
+# Fast/cheap model for bulk chunk summarisation.
+# Accepts any litellm model string: openai/gpt-4o-mini, gemini/gemini-2.0-flash, ollama/llama3, etc.
+# See: https://docs.litellm.ai/docs/providers
+MENTAT_SUMMARY_MODEL=openai/gpt-4o-mini
+# MENTAT_SUMMARY_API_KEY=          # Optional — overrides the provider's global key
+# MENTAT_SUMMARY_API_BASE=         # Optional — custom endpoint (e.g. Azure, vLLM, local proxy)
+
+# --- Embedding Model ---
+# Accepts any litellm embedding model string.
+# Vector dimension is auto-detected from the first embedding call.
+MENTAT_EMBEDDING_MODEL=openai/text-embedding-3-small
+# MENTAT_EMBEDDING_API_KEY=        # Optional — overrides the provider's global key
+# MENTAT_EMBEDDING_API_BASE=       # Optional — custom endpoint
+
+# --- Global Provider API Keys ---
+# litellm reads these natively. Set the one(s) you need.
+# The per-model keys above (MENTAT_SUMMARY_API_KEY, MENTAT_EMBEDDING_API_KEY) take
+# priority when set; otherwise litellm falls back to these.
+OPENAI_API_KEY=sk-...
+# OPENAI_API_BASE=                 # Optional — for OpenAI-compatible endpoints
+# ANTHROPIC_API_KEY=sk-ant-...
+# GEMINI_API_KEY=...
+# AZURE_API_KEY=...
+# AZURE_API_BASE=https://your-resource.openai.azure.com
+# AZURE_API_VERSION=2024-02-01
+# OLLAMA_API_BASE=http://localhost:11434
+
+# --- Storage ---
+MENTAT_DB_PATH=./mentat_db
+MENTAT_STORAGE_DIR=./mentat_files
+
+# --- Background Processing ---
+# Maximum number of documents to process concurrently in the background queue.
+# Higher values = faster batch processing but more memory/API usage.
+# Default: 5
+MENTAT_MAX_CONCURRENT_TASKS=5
+
+# --- Section Heat Tracking ---
+# Exponential decay half-life for section importance scores (seconds).
+# Default: 86400 (24 hours)
+# MENTAT_SECTION_HEAT_HALF_LIFE=86400
+# Score threshold for a section to be considered "hot".
+# Default: 5.0
+# MENTAT_SECTION_HEAT_THRESHOLD=5.0
+# Maximum number of tracked section entries (coldest evicted first).
+# Default: 1000
+# MENTAT_SECTION_HEAT_MAX_ENTRIES=1000
+

mentat_sr-0.1.0/.gitignore

@@ -0,0 +1,33 @@
+# Python
+__pycache__/
+*.pyc
+*.pyo
+*.pyd
+.Python
+env/
+venv/
+.env
+.venv/
+
+# uv
+.uv/
+
+# Mentat Storage
+*mentat_db/
+*mentat_files/
+
+# IDEs
+.vscode/
+.idea/
+
+# Testing
+.pytest_cache/
+.coverage
+htmlcov/
+
+# Mac
+.DS_Store
+
+design-docs/
+samples/
+references/

mentat_sr-0.1.0/.python-version

@@ -0,0 +1 @@
+3.10

mentat_sr-0.1.0/CLAUDE.md

@@ -0,0 +1,116 @@
+# CLAUDE.md
+
+This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
+
+## Project Overview
+
+Mentat is a next-generation Agentic RAG system (Python 3.10+) that transforms "Content Retrieval" into "Strategy Retrieval." Instead of feeding raw documents to an LLM, Mentat uses statistical probes to extract **semantic fingerprints** (hierarchy + metadata + anchors + snippets), then generates template-based actionable reading guides. Small files (< 1000 tokens) bypass skeleton extraction and return full content directly.
+
+## Development Commands
+
+```bash
+# Package manager: uv
+uv sync                              # Install all dependencies
+
+# Configuration
+cp .env.example .env                 # Then edit .env with API keys / model names
+
+# CLI (preferred way to run)
+uv run python -m mentat.cli [COMMAND]
+mentat --debug [COMMAND]             # After install
+
+# FastAPI server (port 7832)
+mentat serve
+
+# Tests (all mocked, no API keys needed; pytest-asyncio with asyncio_mode=auto)
+uv run pytest tests/ -v              # Full suite
+uv run pytest tests/test_smoke.py -v # End-to-end smoke test
+uv run pytest tests/test_queue.py -v # Single test file
+uv run pytest tests/test_smoke.py::test_probe_markdown -v  # Single test
+```
+
+## Architecture
+
+Three-layer unidirectional pipeline: **Probes → Librarian → Storage**
+
+### Layer 1 — Probes (`mentat/probes/`)
+Semantic fingerprinting — no LLM, pure extraction. Each probe implements `BaseProbe` ABC (`can_handle()` + `run()` → `ProbeResult`). Registry in `__init__.py` — 13 probes tried in order, first match wins. Optional-dep probes (Image, DOCX, PPTX, Calendar) degrade gracefully via try/except import.
+
+Key utilities in `_utils.py`: `estimate_tokens` (1 token ≈ 3 chars), `normalize_chunk_sizes` (merges adjacent small chunks <300 tokens, respects H1/H2 boundaries, max merged 1200 tokens), `should_bypass` (<1000 token files return full content).
+
+### Layer 2 — Librarian (`mentat/librarian/`)
+- `engine.py` — Uses `litellm` for all LLM calls. Takes only `ProbeResult` as input — never reads raw files.
+  - **Phase 1 — Chunk Summarisation** (optional): LLM generates 1-3 sentence summaries per chunk (batched, concurrent). Small files bypass this.
+  - **Phase 2 — Instruction Generation** (template-based, no LLM): produces `brief_intro` + `instructions` from ToC + statistics.
+- `instruction_templates.py` — Format-specific reading guide templates used by all probes.
+
+### Layer 3 — Storage (`mentat/storage/`)
+- `BaseVectorStorage` (`base.py`) — ABC interface for vector storage backends. Enables swapping LanceDB for other backends.
+- `LanceDBStorage` (`vector_db.py`) — implements `BaseVectorStorage`. Separate tables for document stubs and chunks. Lazy vector table creation (dimension auto-detected from first embedding). BTREE scalar index on `doc_id` for collection filtering. Chunks include metadata columns (`source`, `indexed_at`, `file_type`, `metadata_json`) for pre-filtering.
+- `MetadataFilter` / `MetadataFilterSet` (`filters.py`) — Backend-agnostic filter representation with SQL builder for LanceDB WHERE clauses. Supports eq, neq, gt, gte, lt, lte, in, like, between operators.
+- `LocalFileStore` (`file_store.py`) — raw file copies for downstream access.
+- `ContentHashCache` (`cache.py`) — SHA-256 deduplication (JSON-backed). Shares `_JsonMap` base class with `PathIndex`.
+- `PathIndex` (`cache.py`) — path→doc_id mapping for file identity tracking. When the same path is re-indexed with changed content, the old document (stubs + chunks) is deleted before creating the new one. Synthetic keys (`__content__:{filename}`) support `add_content()` dedup. Inherits `_JsonMap` base.
+- `CollectionStore` (`collections.py`) — named doc groups as JSON references (no vector duplication).
+
+### Orchestrator (`mentat/core/hub.py`)
+`Mentat` class — singleton via `get_instance()`, reset with `reset()`. Delegates operations to focused modules:
+- `core/indexer.py` — `Indexer`: `add()`, `add_batch()`, `add_content()`, processing status, wait.
+- `core/searcher.py` — `Searcher`: `search()`, `search_grouped()`, `_raw_search()`, result assembly.
+- `core/reader.py` — `Reader`: `inspect()`, `read_segment()`, `read_structured()`, `get_doc_meta()`, `summarize_doc()`.
+
+`MentatConfig` dataclass in `core/models.py` (loads `.env` via python-dotenv, `MENTAT_*` env vars, precedence: explicit arg > env var > default).
+
+**Async processing pipeline** (default): `add()` returns in ~1-3s after probe + stub storage, then queues background embeddings/summarization. Legacy sync: `add(wait=True)` blocks until complete.
+
+### Data Models (`mentat/core/models.py`)
+All Pydantic models and base interfaces in one place: `MentatResult`, `MentatDocResult`, `ChunkResult`, `MentatConfig`, `Collection`, `BaseAdaptor`.
+
+### Background Queue (`mentat/core/queue.py`)
+`ProcessingQueue` + `BackgroundProcessor` — in-memory priority queue (transient, lost on restart). Priority boosting: documents queried before processing completes get +10 priority. Concurrency: `max_concurrent_tasks` (default 5, via `MENTAT_MAX_CONCURRENT_TASKS`).
+
+### Service Layer (`mentat/service.py`)
+Shared stateless functions used by CLI, server, and future SDK: `index_file()`, `search_docs()`, `resolve_doc_id()`, `list_docs()`, etc. CLI formats for terminal; server returns as JSON.
+
+### Other Core Modules
+- `core/embeddings.py` — `BaseEmbedding` ABC + `LiteLLMEmbedding` provider with batching. `EmbeddingRegistry` for pluggable providers. Oversized chunks split with 500-char overlap.
+- `core/access_tracker.py` — two-layer FIFO: recent (LRU) → hot (≥2 accesses). Promotion callback triggers on-demand summarization. Persistent heat map via `heat_map.json` (debounced writes, loaded on init).
+- `core/section_heat.py` — `SectionHeatTracker`: section-level importance tracking with weighted scoring (read_segment=3.0, inspect=2.0, search=1.0), exponential time decay (24h half-life), hot threshold, LRU eviction, and JSON persistence (`section_heat_map.json`). Parent→child propagation via ToC hierarchy.
+- `core/telemetry.py` — context-manager timing for probe/summarize/librarian phases, token savings tracking.
+- `server.py` — FastAPI HTTP server. Delegates to `service.py` for shared operations.
+- `skill.py` — Skill Integration Layer: OpenAI function calling tool schemas + system prompt fragment for agent two-step retrieval protocol. `export_skill()` returns combined payload.
+
+### Public API (`mentat/__init__.py`)
+Thin convenience wrappers delegating to `Mentat` singleton. Module-level async functions: `add()`, `add_batch()`, `add_content()`, `search()`, `search_grouped()`, `inspect()`, `get_doc_meta()`, `read_structured()`, `read_segment()`, `track_access()`, `start_processor()`, `shutdown()`, `wait_for()`. Sync functions: `probe()`, `stats()`, `get_section_heat()`, `collection()`, `collections()`, `get_status()`, `configure()`. Re-exports: `Mentat`, `MentatConfig`, `MentatResult`, `MentatDocResult`, `ChunkResult`, `Collection`, `BaseAdaptor`, `ProbeResult`.
+
+## CLI Commands
+
+```bash
+mentat probe <file_paths> [--format rich|json]        # Run probes (no LLM, no storage)
+mentat index <path> [--force] [-c <collection>] [--summarize] [--llm-instructions] [--wait] [-j N]
+mentat status <doc_id>                                 # Check processing status
+mentat search <query> [--top-k 5] [--hybrid] [-c <collection>] [--toc-only]
+mentat segment <doc_id> <section>                      # Read specific section (two-step protocol step 2)
+mentat inspect <doc_id>                                # Show probe results + instructions
+mentat stats                                           # System statistics
+mentat collection list|show|delete|remove              # Collection management
+mentat skill [--format json|prompt]                     # Export agent tool schemas + system prompt
+mentat serve                                           # Start FastAPI server (port 7832)
+```
+
+## Two-Step Retrieval Protocol (Agent Integration)
+
+Mentat implements a Probe→Fetch protocol for token-efficient agent memory access:
+
+1. **Step 1 — Discover**: `search(query, toc_only=True)` returns document summaries + ToC entries (no chunk content). Agent sees ~100-200 tokens per doc.
+2. **Step 2 — Read**: `read_segment(doc_id, section_path)` fetches specific section content by doc_id + section name from step 1.
+
+The `skill.py` module exports OpenAI function calling tool schemas and a system prompt fragment that teaches agents this protocol. Use `mentat skill` CLI or `GET /skill` endpoint.
+
+## Key Patterns
+
+- **Pydantic v2** for all data models (`ProbeResult`, `TopicInfo`, `StructureInfo`, `Chunk`, `TocEntry`) in `mentat/probes/base.py`, result/config models (`MentatResult`, `MentatDocResult`, `ChunkResult`, `MentatConfig`, `BaseAdaptor`) in `mentat/core/models.py`
+- **Plugin registry** for probes — add new format by implementing `BaseProbe` and registering in `mentat/probes/__init__.py`
+- **Two-stage storage** — stubs stored immediately (with ToC); chunks stored after background embedding/summarization
+- **Collections** — named doc groups for scoped search; shared storage with doc_id references; LanceDB `WHERE doc_id IN (...)` pre-filtering
+- **Config precedence** — explicit arg > `MENTAT_*` env var > default. Separate `api_key`/`api_base` for summary model (`MENTAT_SUMMARY_*`) and embedding model (`MENTAT_EMBEDDING_*`). Global provider keys (`OPENAI_API_KEY`, etc.) read by `litellm` as fallback.

mentat_sr-0.1.0/LICENSE

@@ -0,0 +1,190 @@
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to the Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by the Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding any notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   Copyright 2025 Shelven Zhou
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.