protoprompt 0.1.0__tar.gz

protoprompt-0.1.0/CHANGELOG.md

@@ -0,0 +1,34 @@
+# Changelog
+
+All notable changes to this project will be documented in this file.
+
+The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
+and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
+
+## [0.1.0] - 2026-07-10
+
+### Added
+- `protoprompt.tokens` subpackage with pluggable `TokenCounter` protocol
+  and a regex-based default implementation. Optional `tiktoken` adapter.
+- `LLMSummaryStrategy` for LLM-based session compression with
+  `HeuristicStrategy` fallback on failure.
+- `TokenBudgetedContextBuilder` with priority-based greedy token allocation
+  and `BudgetReport` observability.
+- Real ChromaDB integration tests gated on the `chroma` extra.
+- MkDocs Material documentation under `docs/`.
+- GitHub Actions CI workflow across Python 3.11-3.13.
+
+### Changed
+- `ContextBuilder` now embeds the query once and reuses it for RAG and
+  session retrieval (was: embedded twice).
+- `InMemStore` supports filtering on any metadata field, not only `doc_id`.
+- `StoreProtocol.query` accepts an optional `score_threshold`.
+- `Pipeline.compress_and_store` writes to a `_new` doc_id first, then deletes
+  the old one, to avoid data loss on crash between steps.
+- `ProfileBuilder` logs the original exception instead of silently returning
+  an empty profile.
+- `HeuristicStrategy` honours a `min_messages` threshold.
+
+### Fixed
+- Public re-exports in `protoprompt.profile`, `protoprompt.session`,
+  `protoprompt.store` so IDEs and type checkers see the full API.

protoprompt-0.1.0/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 EnergoAI Hub
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

protoprompt-0.1.0/MANIFEST.in

@@ -0,0 +1,11 @@
+include README.md
+include README.en.md
+include README.ru.md
+include LICENSE
+include CHANGELOG.md
+recursive-include docs *.md
+recursive-include scripts *.py
+recursive-include examples *.py
+recursive-include tests *.py
+include mkdocs.ru.yml
+include mkdocs.en.yml

protoprompt-0.1.0/PKG-INFO

@@ -0,0 +1,201 @@
+Metadata-Version: 2.4
+Name: protoprompt
+Version: 0.1.0
+Summary: Layered context builder for LLM prompts: RAG + compressed session memory + user profile
+Author: EnergoAI Hub
+Maintainer: EnergoAI Hub
+License: MIT
+Project-URL: Homepage, https://github.com/Idxeed/protoprompt
+Project-URL: Documentation, https://idxeed.github.io/protoprompt/
+Project-URL: Source, https://github.com/Idxeed/protoprompt
+Project-URL: Issues, https://github.com/Idxeed/protoprompt/issues
+Project-URL: Changelog, https://github.com/Idxeed/protoprompt/blob/main/CHANGELOG.md
+Keywords: llm,rag,prompt,context,embedding
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Developers
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
+Classifier: Typing :: Typed
+Requires-Python: >=3.11
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Provides-Extra: chroma
+Requires-Dist: chromadb<0.6,>=0.5; extra == "chroma"
+Provides-Extra: tiktoken
+Requires-Dist: tiktoken>=0.5; extra == "tiktoken"
+Provides-Extra: dev
+Requires-Dist: pytest>=8; extra == "dev"
+Requires-Dist: pytest-asyncio>=0.24; extra == "dev"
+Requires-Dist: pytest-cov>=5; extra == "dev"
+Requires-Dist: mkdocs>=1.5; extra == "dev"
+Requires-Dist: mkdocs-material>=9; extra == "dev"
+Requires-Dist: mkdocstrings[python]>=0.24; extra == "dev"
+Dynamic: license-file
+
+# protoprompt
+
+[![CI](https://github.com/Idxeed/protoprompt/actions/workflows/ci.yml/badge.svg)](https://github.com/Idxeed/protoprompt/actions/workflows/ci.yml)
+[![Coverage](https://img.shields.io/codecov/c/github/Idxeed/protoprompt)](https://codecov.io/gh/Idxeed/protoprompt)
+[![Python 3.11+](https://img.shields.io/badge/python-3.11%2B-blue)](https://www.python.org/)
+[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)
+[![RU](https://img.shields.io/badge/%D0%AF%D0%B7%D1%8B%D0%BA-RU-blue)](README.ru.md)
+[![EN](https://img.shields.io/badge/Language-EN-blue)](README.en.md)
+
+Слоистый сборщик контекста для LLM-промптов. Три независимых, компонуемых
+слоя кормят модель: **RAG по документам**, **сжатая история сессии** и
+опциональный **профиль пользователя**. Подключаемое векторное хранилище,
+подключаемый токенайзер, подключаемая стратегия сжатия.
+
+[English version](README.en.md)
+
+## Зачем
+
+Прод-LLM-приложения упираются в одну и ту же стену: модели нужен контекст,
+но окно конечно. Собранная вручную сборка промпта превращается в кашу:
+документы ищутся отдельно от истории, system prompt дублируется, и каждая
+команда переписывает один и тот же клей.
+
+`protoprompt` разделяет три задачи и даёт каждой чистый протокол:
+
+- `StoreProtocol` — векторное хранилище (in-memory для тестов, ChromaDB для прода).
+- `StrategyProtocol` — как сжимать старые ходы длинной сессии.
+- `TokenCounter` — как считать бюджет финального промпта.
+
+`ContextBuilder` оркестрирует все три. Результат — единый `system_prompt`
+плюс структурированный `ContextOutput` с описанием того, что вошло (head,
+tail, RAG, profile), чтобы UI мог показать provenance.
+
+## Установка
+
+```bash
+pip install protoprompt
+
+# С ChromaDB-бэкендом
+pip install "protoprompt[chroma]"
+
+# С tiktoken-токенайзером
+pip install "protoprompt[tiktoken]"
+
+# Для разработки и документации
+pip install "protoprompt[chroma,dev]"
+```
+
+## Быстрый старт
+
+```python
+import asyncio
+from protoprompt import (
+    InMemStore,
+    ContextBuilder,
+    ContextInput,
+    Pipeline,
+    HeuristicStrategy,
+    Session,
+)
+
+
+class MyLLM:
+    async def chat(self, messages, model="", **options):
+        return "заглушка"
+
+    async def embed(self, texts, model=""):
+        # замените на реальные эмбеддинги
+        return [[0.1] * 384 for _ in texts]
+
+
+async def main():
+    store = InMemStore()
+    store.add("doc-1", ["Париж — столица Франции."], [[0.5] * 384])
+    llm = MyLLM()
+
+    builder = ContextBuilder(store, llm)
+    out = await builder.build(ContextInput(
+        query="Какая столица Франции?",
+        system_prompt="Ты учитель географии.",
+        doc_ids=[1],
+    ))
+    print(out.system_prompt)
+
+    pipeline = Pipeline(
+        store, llm,
+        strategy=HeuristicStrategy(),
+        compress_every_n=10,
+    )
+    session = Session(chat_id="c1", messages=[
+        {"role": "user", "content": "Привет"},
+        {"role": "assistant", "content": "Здравствуйте!"},
+    ])
+    if pipeline.should_compress(len(session.messages)):
+        await pipeline.compress_and_store(session)
+
+
+asyncio.run(main())
+```
+
+## Архитектура
+
+```
+                 +--------------------+
+                 |   ContextBuilder   |
+                 |   (оркестратор)    |
+                 +---------+----------+
+                           |
+        +------------------+------------------+------------------+
+        |                  |                  |                  |
+        v                  v                  v                  v
++---------------+  +----------------+  +---------------+  +---------+
+| RAG-поиск     |  | Память сессии  |  | Профиль польз.|  |  Store  |
+| (vector top-k)|  | (сжатая)       |  | (из LLM)      |  |  query  |
++---------------+  +----------------+  +---------------+  +---------+
+        |                  |                  |
+        +--------+---------+--------+---------+
+                 |                  |
+                 v                  v
+        +----------------+  +----------------+
+        |  TokenCounter  |  |  StoreProtocol |
+        |  (pluggable)   |  |  (in-mem/chroma)|
+        +----------------+  +----------------+
+```
+
+## Публичный API
+
+| Модуль                   | Экспорты                                                            |
+|--------------------------|---------------------------------------------------------------------|
+| `protoprompt`            | `Pipeline`, `ContextBuilder`, `ContextInput`, `ContextOutput`      |
+| `protoprompt.store`      | `StoreProtocol`, `InMemStore`, `ChromaStore`                        |
+| `protoprompt.session`    | `Session`, `CompressedBlock`, `HeuristicStrategy`, `LLMSummaryStrategy` |
+| `protoprompt.profile`    | `UserProfile`, `ProfileBuilder`                                     |
+| `protoprompt.tokens`     | `TokenCounter`, `RegexTokenCounter`, `TiktokenCounter`              |
+| `protoprompt.llm`        | `LLMClientProtocol`                                                 |
+
+## Документация
+
+Полная документация собирается в двух языковых версиях:
+
+- 🇷🇺 Русская: <https://idxeed.github.io/protoprompt/ru/>
+- 🇬🇧 English: <https://idxeed.github.io/protoprompt/en/>
+
+Локальная сборка:
+
+```bash
+pip install "protoprompt[dev]"
+python scripts/build_docs.py --serve  # обе версии на разных портах
+```
+
+## Разработка
+
+```bash
+git clone https://github.com/Idxeed/protoprompt
+cd protoprompt
+python -m venv .venv && source .venv/bin/activate
+pip install -e ".[chroma,dev]"
+pytest
+```
+
+## Лицензия
+
+MIT — см. [LICENSE](LICENSE).

protoprompt-0.1.0/README.en.md

@@ -0,0 +1,164 @@
+# protoprompt
+
+[![CI](https://github.com/Idxeed/protoprompt/actions/workflows/ci.yml/badge.svg)](https://github.com/Idxeed/protoprompt/actions/workflows/ci.yml)
+[![Coverage](https://img.shields.io/codecov/c/github/Idxeed/protoprompt)](https://codecov.io/gh/Idxeed/protoprompt)
+[![Python 3.11+](https://img.shields.io/badge/python-3.11%2B-blue)](https://www.python.org/)
+[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)
+[![RU](https://img.shields.io/badge/%D0%AF%D0%B7%D1%8B%D0%BA-RU-blue)](README.ru.md)
+[![EN](https://img.shields.io/badge/Language-EN-blue)](README.en.md)
+
+[Русская версия](README.ru.md)
+
+Layered context builder for LLM prompts. Three independent, composable
+layers feed the model: **RAG over documents**, **compressed session
+memory**, and an optional **user profile**. Pluggable vector store,
+pluggable tokenizer, pluggable compression strategy.
+
+## Why
+
+Production LLM apps hit the same wall: the model needs context, but the
+context window is finite. Hand-rolled prompt assembly gets messy:
+documents are queried separately from session history, the system prompt
+gets duplicated, and you end up rewriting the same glue code per project.
+
+`protoprompt` separates the three concerns and gives each a clean
+protocol:
+
+- `StoreProtocol` — vector storage (in-memory for tests, ChromaDB for prod).
+- `StrategyProtocol` — how to compress old turns of a long session.
+- `TokenCounter` — how to budget the final prompt.
+
+The `ContextBuilder` orchestrates all three. The result is a single
+`system_prompt` plus a structured `ContextOutput` describing what went in
+(head, tail, RAG, profile) so the UI can show provenance.
+
+## Install
+
+```bash
+pip install protoprompt
+
+# With ChromaDB backend
+pip install "protoprompt[chroma]"
+
+# With tiktoken-based tokenizer
+pip install "protoprompt[tiktoken]"
+
+# Dev / docs
+pip install "protoprompt[chroma,dev]"
+```
+
+## Quickstart
+
+```python
+import asyncio
+from protoprompt import (
+    InMemStore,
+    ContextBuilder,
+    ContextInput,
+    Pipeline,
+    HeuristicStrategy,
+    Session,
+)
+
+
+class MyLLM:
+    async def chat(self, messages, model="", **options):
+        return "stub"
+
+    async def embed(self, texts, model=""):
+        # replace with real embeddings
+        return [[0.1] * 384 for _ in texts]
+
+
+async def main():
+    store = InMemStore()
+    store.add("doc-1", ["Paris is the capital of France."], [[0.5] * 384])
+    llm = MyLLM()
+
+    builder = ContextBuilder(store, llm)
+    out = await builder.build(ContextInput(
+        query="What is the capital of France?",
+        system_prompt="You are a geography tutor.",
+        doc_ids=[1],
+    ))
+    print(out.system_prompt)
+
+    pipeline = Pipeline(
+        store, llm,
+        strategy=HeuristicStrategy(),
+        compress_every_n=10,
+    )
+    session = Session(chat_id="c1", messages=[
+        {"role": "user", "content": "Hi"},
+        {"role": "assistant", "content": "Hello!"},
+    ])
+    if pipeline.should_compress(len(session.messages)):
+        await pipeline.compress_and_store(session)
+
+
+asyncio.run(main())
+```
+
+## Architecture
+
+```
+                 +--------------------+
+                 |   ContextBuilder   |
+                 |   (orchestrator)   |
+                 +---------+----------+
+                           |
+        +------------------+------------------+------------------+
+        |                  |                  |                  |
+        v                  v                  v                  v
++---------------+  +----------------+  +---------------+  +---------+
+| RAG retrieval |  | Session memory |  | User profile  |  |  Store  |
+| (vector top-k)|  | (compressed)   |  | (LLM-derived) |  |  query  |
++---------------+  +----------------+  +---------------+  +---------+
+        |                  |                  |
+        +--------+---------+--------+---------+
+                 |                  |
+                 v                  v
+        +----------------+  +----------------+
+        |  TokenCounter  |  |  StoreProtocol |
+        |  (pluggable)   |  |  (in-mem/chroma)|
+        +----------------+  +----------------+
+```
+
+## Public API
+
+| Module                   | Exports                                                           |
+|--------------------------|-------------------------------------------------------------------|
+| `protoprompt`            | `Pipeline`, `ContextBuilder`, `ContextInput`, `ContextOutput`     |
+| `protoprompt.store`      | `StoreProtocol`, `InMemStore`, `ChromaStore`                      |
+| `protoprompt.session`    | `Session`, `CompressedBlock`, `HeuristicStrategy`, `LLMSummaryStrategy` |
+| `protoprompt.profile`    | `UserProfile`, `ProfileBuilder`                                   |
+| `protoprompt.tokens`     | `TokenCounter`, `RegexTokenCounter`, `TiktokenCounter`            |
+| `protoprompt.llm`        | `LLMClientProtocol`                                               |
+
+## Documentation
+
+Full docs are built in two languages:
+
+- 🇷🇺 Russian: <https://idxeed.github.io/protoprompt/ru/>
+- 🇬🇧 English: <https://idxeed.github.io/protoprompt/en/>
+
+Build locally:
+
+```bash
+pip install "protoprompt[dev]"
+python scripts/build_docs.py --serve  # both versions on different ports
+```
+
+## Development
+
+```bash
+git clone https://github.com/Idxeed/protoprompt
+cd protoprompt
+python -m venv .venv && source .venv/bin/activate
+pip install -e ".[chroma,dev]"
+pytest
+```
+
+## License
+
+MIT — see [LICENSE](LICENSE).

protoprompt-0.1.0/README.md

@@ -0,0 +1,163 @@
+# protoprompt
+
+[![CI](https://github.com/Idxeed/protoprompt/actions/workflows/ci.yml/badge.svg)](https://github.com/Idxeed/protoprompt/actions/workflows/ci.yml)
+[![Coverage](https://img.shields.io/codecov/c/github/Idxeed/protoprompt)](https://codecov.io/gh/Idxeed/protoprompt)
+[![Python 3.11+](https://img.shields.io/badge/python-3.11%2B-blue)](https://www.python.org/)
+[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)
+[![RU](https://img.shields.io/badge/%D0%AF%D0%B7%D1%8B%D0%BA-RU-blue)](README.ru.md)
+[![EN](https://img.shields.io/badge/Language-EN-blue)](README.en.md)
+
+Слоистый сборщик контекста для LLM-промптов. Три независимых, компонуемых
+слоя кормят модель: **RAG по документам**, **сжатая история сессии** и
+опциональный **профиль пользователя**. Подключаемое векторное хранилище,
+подключаемый токенайзер, подключаемая стратегия сжатия.
+
+[English version](README.en.md)
+
+## Зачем
+
+Прод-LLM-приложения упираются в одну и ту же стену: модели нужен контекст,
+но окно конечно. Собранная вручную сборка промпта превращается в кашу:
+документы ищутся отдельно от истории, system prompt дублируется, и каждая
+команда переписывает один и тот же клей.
+
+`protoprompt` разделяет три задачи и даёт каждой чистый протокол:
+
+- `StoreProtocol` — векторное хранилище (in-memory для тестов, ChromaDB для прода).
+- `StrategyProtocol` — как сжимать старые ходы длинной сессии.
+- `TokenCounter` — как считать бюджет финального промпта.
+
+`ContextBuilder` оркестрирует все три. Результат — единый `system_prompt`
+плюс структурированный `ContextOutput` с описанием того, что вошло (head,
+tail, RAG, profile), чтобы UI мог показать provenance.
+
+## Установка
+
+```bash
+pip install protoprompt
+
+# С ChromaDB-бэкендом
+pip install "protoprompt[chroma]"
+
+# С tiktoken-токенайзером
+pip install "protoprompt[tiktoken]"
+
+# Для разработки и документации
+pip install "protoprompt[chroma,dev]"
+```
+
+## Быстрый старт
+
+```python
+import asyncio
+from protoprompt import (
+    InMemStore,
+    ContextBuilder,
+    ContextInput,
+    Pipeline,
+    HeuristicStrategy,
+    Session,
+)
+
+
+class MyLLM:
+    async def chat(self, messages, model="", **options):
+        return "заглушка"
+
+    async def embed(self, texts, model=""):
+        # замените на реальные эмбеддинги
+        return [[0.1] * 384 for _ in texts]
+
+
+async def main():
+    store = InMemStore()
+    store.add("doc-1", ["Париж — столица Франции."], [[0.5] * 384])
+    llm = MyLLM()
+
+    builder = ContextBuilder(store, llm)
+    out = await builder.build(ContextInput(
+        query="Какая столица Франции?",
+        system_prompt="Ты учитель географии.",
+        doc_ids=[1],
+    ))
+    print(out.system_prompt)
+
+    pipeline = Pipeline(
+        store, llm,
+        strategy=HeuristicStrategy(),
+        compress_every_n=10,
+    )
+    session = Session(chat_id="c1", messages=[
+        {"role": "user", "content": "Привет"},
+        {"role": "assistant", "content": "Здравствуйте!"},
+    ])
+    if pipeline.should_compress(len(session.messages)):
+        await pipeline.compress_and_store(session)
+
+
+asyncio.run(main())
+```
+
+## Архитектура
+
+```
+                 +--------------------+
+                 |   ContextBuilder   |
+                 |   (оркестратор)    |
+                 +---------+----------+
+                           |
+        +------------------+------------------+------------------+
+        |                  |                  |                  |
+        v                  v                  v                  v
++---------------+  +----------------+  +---------------+  +---------+
+| RAG-поиск     |  | Память сессии  |  | Профиль польз.|  |  Store  |
+| (vector top-k)|  | (сжатая)       |  | (из LLM)      |  |  query  |
++---------------+  +----------------+  +---------------+  +---------+
+        |                  |                  |
+        +--------+---------+--------+---------+
+                 |                  |
+                 v                  v
+        +----------------+  +----------------+
+        |  TokenCounter  |  |  StoreProtocol |
+        |  (pluggable)   |  |  (in-mem/chroma)|
+        +----------------+  +----------------+
+```
+
+## Публичный API
+
+| Модуль                   | Экспорты                                                            |
+|--------------------------|---------------------------------------------------------------------|
+| `protoprompt`            | `Pipeline`, `ContextBuilder`, `ContextInput`, `ContextOutput`      |
+| `protoprompt.store`      | `StoreProtocol`, `InMemStore`, `ChromaStore`                        |
+| `protoprompt.session`    | `Session`, `CompressedBlock`, `HeuristicStrategy`, `LLMSummaryStrategy` |
+| `protoprompt.profile`    | `UserProfile`, `ProfileBuilder`                                     |
+| `protoprompt.tokens`     | `TokenCounter`, `RegexTokenCounter`, `TiktokenCounter`              |
+| `protoprompt.llm`        | `LLMClientProtocol`                                                 |
+
+## Документация
+
+Полная документация собирается в двух языковых версиях:
+
+- 🇷🇺 Русская: <https://idxeed.github.io/protoprompt/ru/>
+- 🇬🇧 English: <https://idxeed.github.io/protoprompt/en/>
+
+Локальная сборка:
+
+```bash
+pip install "protoprompt[dev]"
+python scripts/build_docs.py --serve  # обе версии на разных портах
+```
+
+## Разработка
+
+```bash
+git clone https://github.com/Idxeed/protoprompt
+cd protoprompt
+python -m venv .venv && source .venv/bin/activate
+pip install -e ".[chroma,dev]"
+pytest
+```
+
+## Лицензия
+
+MIT — см. [LICENSE](LICENSE).

protoprompt-0.1.0/docs/en/api/context.md

@@ -0,0 +1,3 @@
+# API: context
+
+::: protoprompt.context

protoprompt-0.1.0/docs/en/api/injector.md

@@ -0,0 +1,17 @@
+# API: injector
+
+## ContextBuilder
+
+::: protoprompt.injector.ContextBuilder
+
+## TokenBudgetedContextBuilder
+
+::: protoprompt.injector_budgeted.TokenBudgetedContextBuilder
+
+## BudgetReport
+
+::: protoprompt.injector_budgeted.BudgetReport
+
+## Exceptions
+
+::: protoprompt.exceptions

protoprompt-0.1.0/docs/en/api/pipeline.md

@@ -0,0 +1,5 @@
+# API: pipeline
+
+## Pipeline
+
+::: protoprompt.pipeline.Pipeline

protoprompt-0.1.0/docs/en/api/profile.md

@@ -0,0 +1,9 @@
+# API: profile
+
+## UserProfile
+
+::: protoprompt.profile.types
+
+## ProfileBuilder
+
+::: protoprompt.profile.builder.ProfileBuilder

protoprompt-0.1.0/docs/en/api/session.md

@@ -0,0 +1,17 @@
+# API: session
+
+## Session & CompressedBlock
+
+::: protoprompt.session.types
+
+## HeuristicStrategy
+
+::: protoprompt.session.strategy.HeuristicStrategy
+
+## LLMSummaryStrategy
+
+::: protoprompt.session.strategy.LLMSummaryStrategy
+
+## Compressor
+
+::: protoprompt.session.compressor.Compressor

protoprompt-0.1.0/docs/en/api/store.md

@@ -0,0 +1,13 @@
+# API: store
+
+## StoreProtocol
+
+::: protoprompt.store.protocol
+
+## InMemStore
+
+::: protoprompt.store.memory.InMemStore
+
+## ChromaStore
+
+::: protoprompt.store.chroma.ChromaStore

protoprompt-0.1.0/docs/en/api/tokens.md

@@ -0,0 +1,13 @@
+# API: tokens
+
+## TokenCounter
+
+::: protoprompt.tokens.protocol
+
+## RegexTokenCounter
+
+::: protoprompt.tokens.regex_counter
+
+## TiktokenCounter
+
+::: protoprompt.tokens.tiktoken_adapter