agentcamp 0.1.0__tar.gz

agentcamp-0.1.0/CHANGELOG.md

@@ -0,0 +1,18 @@
+# Changelog
+
+All notable changes to the `agentcamp` package will be documented in this file.
+
+The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
+and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
+
+## [Unreleased]
+
+## [0.1.0] - 2026-06-01
+
+### Added
+- Initial PyPI release of the `agentcamp` package.
+- `AgentCampClient` with `project`, `api_key`, `mode`, and `base_url` configuration.
+- `wrap_langgraph(graph, project=...)` convenience entrypoint (no-op passthrough for now).
+- `Mode` enum: `OBSERVE`, `TRAIN`, `SHADOW`, `ACTIVATE`.
+- `py.typed` marker for type-checker support.
+- PyPI packaging + release infrastructure (GitHub Actions, release scripts, MANIFEST).

agentcamp-0.1.0/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 FintorAI
+
+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.

agentcamp-0.1.0/MANIFEST.in

@@ -0,0 +1,8 @@
+include LICENSE
+include README.md
+include CHANGELOG.md
+include src/agentcamp/py.typed
+
+recursive-exclude * __pycache__
+recursive-exclude * *.py[co]
+recursive-exclude * .DS_Store

agentcamp-0.1.0/PKG-INFO

@@ -0,0 +1,96 @@
+Metadata-Version: 2.4
+Name: agentcamp
+Version: 0.1.0
+Summary: AgentCamp SDK — where agents train their own experts. Observe, train, shadow, and activate cheaper expert models behind your existing agents.
+Author: FintorAI
+License-Expression: MIT
+Project-URL: Homepage, https://agentcamp.ai
+Project-URL: Repository, https://github.com/FintorAI/agentcamp
+Project-URL: Changelog, https://github.com/FintorAI/agentcamp/blob/main/CHANGELOG.md
+Keywords: agents,langgraph,llm,fine-tuning,agentcamp,distillation
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Developers
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.9
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Typing :: Typed
+Requires-Python: <4.0,>=3.9
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Provides-Extra: dev
+Requires-Dist: pytest; extra == "dev"
+Requires-Dist: pytest-cov; extra == "dev"
+Requires-Dist: build; extra == "dev"
+Requires-Dist: twine; extra == "dev"
+Dynamic: license-file
+
+# AgentCamp
+
+[![PyPI version](https://img.shields.io/pypi/v/agentcamp.svg)](https://pypi.org/project/agentcamp/)
+[![Python versions](https://img.shields.io/pypi/pyversions/agentcamp.svg)](https://pypi.org/project/agentcamp/)
+
+**Where agents train their own experts. Same results, dramatically cheaper.**
+
+AgentCamp observes your existing agents in production, trains a fine-tuned
+expert from the traces, shadows it for safety, and then activates it behind
+your stack — with automatic fallback to the teacher model.
+
+> v0.1.0 is an early placeholder release. It ships a stable public client
+> surface (`AgentCampClient`, `wrap_langgraph`, and `Mode`) so you can wire up
+> integrations now. Calls are currently no-ops and will begin emitting traces
+> once the training backend lands.
+
+## Installation
+
+```bash
+pip install agentcamp
+```
+
+## Quick start
+
+```python
+import agentcamp
+
+# Wrap an existing LangGraph graph so AgentCamp can observe its runs.
+graph = agentcamp.wrap_langgraph(graph, project="support-agent")
+```
+
+Or use the client directly:
+
+```python
+from agentcamp import AgentCampClient, Mode
+
+client = AgentCampClient(project="support-agent", mode=Mode.OBSERVE)
+graph = client.wrap_langgraph(graph)
+```
+
+## Modes
+
+| Mode       | What it does                                                        |
+| ---------- | ------------------------------------------------------------------- |
+| `OBSERVE`  | Collect traces from the teacher agent only.                         |
+| `TRAIN`    | Build a fine-tuned expert from collected traces.                    |
+| `SHADOW`   | Run the expert alongside the teacher without serving it.            |
+| `ACTIVATE` | Serve the expert, falling back to the teacher when needed.          |
+
+## Configuration
+
+| Environment variable | Purpose                                  |
+| -------------------- | ---------------------------------------- |
+| `AGENTCAMP_API_KEY`  | API key used to authenticate the client. |
+| `AGENTCAMP_BASE_URL` | Override the AgentCamp API base URL.     |
+
+## Development
+
+```bash
+python -m venv .venv
+source .venv/bin/activate
+pip install -e '.[dev]'
+pytest
+```
+
+## License
+
+MIT — see [LICENSE](LICENSE).

agentcamp-0.1.0/README.md

@@ -0,0 +1,68 @@
+# AgentCamp
+
+[![PyPI version](https://img.shields.io/pypi/v/agentcamp.svg)](https://pypi.org/project/agentcamp/)
+[![Python versions](https://img.shields.io/pypi/pyversions/agentcamp.svg)](https://pypi.org/project/agentcamp/)
+
+**Where agents train their own experts. Same results, dramatically cheaper.**
+
+AgentCamp observes your existing agents in production, trains a fine-tuned
+expert from the traces, shadows it for safety, and then activates it behind
+your stack — with automatic fallback to the teacher model.
+
+> v0.1.0 is an early placeholder release. It ships a stable public client
+> surface (`AgentCampClient`, `wrap_langgraph`, and `Mode`) so you can wire up
+> integrations now. Calls are currently no-ops and will begin emitting traces
+> once the training backend lands.
+
+## Installation
+
+```bash
+pip install agentcamp
+```
+
+## Quick start
+
+```python
+import agentcamp
+
+# Wrap an existing LangGraph graph so AgentCamp can observe its runs.
+graph = agentcamp.wrap_langgraph(graph, project="support-agent")
+```
+
+Or use the client directly:
+
+```python
+from agentcamp import AgentCampClient, Mode
+
+client = AgentCampClient(project="support-agent", mode=Mode.OBSERVE)
+graph = client.wrap_langgraph(graph)
+```
+
+## Modes
+
+| Mode       | What it does                                                        |
+| ---------- | ------------------------------------------------------------------- |
+| `OBSERVE`  | Collect traces from the teacher agent only.                         |
+| `TRAIN`    | Build a fine-tuned expert from collected traces.                    |
+| `SHADOW`   | Run the expert alongside the teacher without serving it.            |
+| `ACTIVATE` | Serve the expert, falling back to the teacher when needed.          |
+
+## Configuration
+
+| Environment variable | Purpose                                  |
+| -------------------- | ---------------------------------------- |
+| `AGENTCAMP_API_KEY`  | API key used to authenticate the client. |
+| `AGENTCAMP_BASE_URL` | Override the AgentCamp API base URL.     |
+
+## Development
+
+```bash
+python -m venv .venv
+source .venv/bin/activate
+pip install -e '.[dev]'
+pytest
+```
+
+## License
+
+MIT — see [LICENSE](LICENSE).

agentcamp-0.1.0/pyproject.toml

@@ -0,0 +1,85 @@
+[project]
+name = "agentcamp"
+version = "0.1.0"
+description = "AgentCamp SDK — where agents train their own experts. Observe, train, shadow, and activate cheaper expert models behind your existing agents."
+readme = "README.md"
+license = "MIT"
+requires-python = ">=3.9,<4.0"
+authors = [
+    { name = "FintorAI" },
+]
+keywords = ["agents", "langgraph", "llm", "fine-tuning", "agentcamp", "distillation"]
+classifiers = [
+    "Development Status :: 3 - Alpha",
+    "Intended Audience :: Developers",
+    "Programming Language :: Python :: 3",
+    "Programming Language :: Python :: 3.9",
+    "Programming Language :: Python :: 3.10",
+    "Programming Language :: Python :: 3.11",
+    "Programming Language :: Python :: 3.12",
+    "Typing :: Typed",
+]
+dependencies = []
+
+[project.optional-dependencies]
+dev = [
+    "pytest",
+    "pytest-cov",
+    "build",
+    "twine",
+]
+
+[project.urls]
+Homepage = "https://agentcamp.ai"
+Repository = "https://github.com/FintorAI/agentcamp"
+Changelog = "https://github.com/FintorAI/agentcamp/blob/main/CHANGELOG.md"
+
+[dependency-groups]
+lint = [
+    "ruff>=0.12.2,<0.13.0",
+]
+
+[build-system]
+requires = ["setuptools>=73.0.0", "wheel"]
+build-backend = "setuptools.build_meta"
+
+[tool.setuptools]
+[tool.setuptools.packages.find]
+where = ["src"]
+include = ["agentcamp*"]
+
+[tool.setuptools.package-dir]
+"" = "src"
+
+[tool.setuptools.package-data]
+"*" = ["py.typed"]
+
+[tool.ruff]
+line-length = 150
+exclude = []
+
+[tool.ruff.lint]
+select = ["E", "F", "I", "UP", "B"]
+ignore = [
+    "COM812",
+    "ISC001",
+]
+
+[tool.pytest.ini_options]
+pythonpath = ["src"]
+testpaths = ["tests"]
+
+[tool.bumpversion]
+current_version = "0.1.0"
+commit = true
+tag = true
+
+[[tool.bumpversion.files]]
+filename = "pyproject.toml"
+search = 'version = "{current_version}"'
+replace = 'version = "{new_version}"'
+
+[[tool.bumpversion.files]]
+filename = "src/agentcamp/__init__.py"
+search = '__version__ = "{current_version}"'
+replace = '__version__ = "{new_version}"'

agentcamp-0.1.0/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+

agentcamp-0.1.0/src/agentcamp/__init__.py

@@ -0,0 +1,24 @@
+"""AgentCamp SDK.
+
+Where agents train their own experts. This is the public client surface for
+AgentCamp. v0.1.0 ships a lightweight client stub plus the ``wrap_langgraph``
+entrypoint so integrations can be wired up before the training backend lands.
+"""
+
+__author__ = "FintorAI"
+__version__ = "0.1.0"
+
+from agentcamp.client import (
+    AgentCampClient,
+    AgentCampError,
+    Mode,
+    wrap_langgraph,
+)
+
+__all__ = [
+    "AgentCampClient",
+    "AgentCampError",
+    "Mode",
+    "__version__",
+    "wrap_langgraph",
+]

agentcamp-0.1.0/src/agentcamp/client.py

@@ -0,0 +1,96 @@
+"""Core AgentCamp client.
+
+This is the v0.1.0 placeholder surface. It establishes the public API shape
+(client + ``wrap_langgraph`` + modes) without depending on the training
+backend yet, so downstream code can integrate against a stable import path.
+Calls are no-ops today and will start emitting traces once the backend is wired.
+"""
+
+from __future__ import annotations
+
+import os
+from enum import Enum
+from typing import Any, Optional
+
+__all__ = [
+    "AgentCampClient",
+    "AgentCampError",
+    "Mode",
+    "wrap_langgraph",
+]
+
+
+class AgentCampError(Exception):
+    """Base error for all AgentCamp client failures."""
+
+
+class Mode(str, Enum):
+    """AgentCamp operating modes for a wrapped agent.
+
+    OBSERVE   collect traces from the teacher agent only.
+    TRAIN     build a fine-tuned expert from collected traces.
+    SHADOW    run the expert alongside the teacher without serving it.
+    ACTIVATE  serve the expert, falling back to the teacher when needed.
+    """
+
+    OBSERVE = "observe"
+    TRAIN = "train"
+    SHADOW = "shadow"
+    ACTIVATE = "activate"
+
+
+class AgentCampClient:
+    """Lightweight AgentCamp client.
+
+    Args:
+        project: Project slug that traces and experts are grouped under.
+        api_key: AgentCamp API key. Falls back to ``AGENTCAMP_API_KEY``.
+        mode: Operating mode for this client. Defaults to ``Mode.OBSERVE``.
+        base_url: AgentCamp API base URL. Falls back to ``AGENTCAMP_BASE_URL``.
+    """
+
+    def __init__(
+        self,
+        project: str,
+        api_key: Optional[str] = None,
+        mode: Mode = Mode.OBSERVE,
+        base_url: Optional[str] = None,
+    ) -> None:
+        if not project:
+            raise AgentCampError("`project` is required.")
+        self.project = project
+        self.api_key = api_key or os.getenv("AGENTCAMP_API_KEY")
+        self.mode = Mode(mode)
+        self.base_url = base_url or os.getenv(
+            "AGENTCAMP_BASE_URL", "https://api.agentcamp.ai"
+        )
+
+    def __repr__(self) -> str:
+        return (
+            f"AgentCampClient(project={self.project!r}, mode={self.mode.value!r})"
+        )
+
+    def wrap_langgraph(self, graph: Any, mode: Optional[Mode] = None) -> Any:
+        """Wrap a LangGraph graph so AgentCamp can observe/train on its runs.
+
+        v0.1.0 is a no-op that returns the graph unchanged so integrations can
+        be wired ahead of the training backend.
+        """
+        _ = mode or self.mode
+        return graph
+
+
+def wrap_langgraph(
+    graph: Any,
+    project: str,
+    api_key: Optional[str] = None,
+    mode: Mode = Mode.OBSERVE,
+) -> Any:
+    """Convenience wrapper around :class:`AgentCampClient`.
+
+    Example:
+        >>> import agentcamp
+        >>> graph = agentcamp.wrap_langgraph(graph, project="support-agent")
+    """
+    client = AgentCampClient(project=project, api_key=api_key, mode=mode)
+    return client.wrap_langgraph(graph)

agentcamp-0.1.0/src/agentcamp.egg-info/PKG-INFO

@@ -0,0 +1,96 @@
+Metadata-Version: 2.4
+Name: agentcamp
+Version: 0.1.0
+Summary: AgentCamp SDK — where agents train their own experts. Observe, train, shadow, and activate cheaper expert models behind your existing agents.
+Author: FintorAI
+License-Expression: MIT
+Project-URL: Homepage, https://agentcamp.ai
+Project-URL: Repository, https://github.com/FintorAI/agentcamp
+Project-URL: Changelog, https://github.com/FintorAI/agentcamp/blob/main/CHANGELOG.md
+Keywords: agents,langgraph,llm,fine-tuning,agentcamp,distillation
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Developers
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.9
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Typing :: Typed
+Requires-Python: <4.0,>=3.9
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Provides-Extra: dev
+Requires-Dist: pytest; extra == "dev"
+Requires-Dist: pytest-cov; extra == "dev"
+Requires-Dist: build; extra == "dev"
+Requires-Dist: twine; extra == "dev"
+Dynamic: license-file
+
+# AgentCamp
+
+[![PyPI version](https://img.shields.io/pypi/v/agentcamp.svg)](https://pypi.org/project/agentcamp/)
+[![Python versions](https://img.shields.io/pypi/pyversions/agentcamp.svg)](https://pypi.org/project/agentcamp/)
+
+**Where agents train their own experts. Same results, dramatically cheaper.**
+
+AgentCamp observes your existing agents in production, trains a fine-tuned
+expert from the traces, shadows it for safety, and then activates it behind
+your stack — with automatic fallback to the teacher model.
+
+> v0.1.0 is an early placeholder release. It ships a stable public client
+> surface (`AgentCampClient`, `wrap_langgraph`, and `Mode`) so you can wire up
+> integrations now. Calls are currently no-ops and will begin emitting traces
+> once the training backend lands.
+
+## Installation
+
+```bash
+pip install agentcamp
+```
+
+## Quick start
+
+```python
+import agentcamp
+
+# Wrap an existing LangGraph graph so AgentCamp can observe its runs.
+graph = agentcamp.wrap_langgraph(graph, project="support-agent")
+```
+
+Or use the client directly:
+
+```python
+from agentcamp import AgentCampClient, Mode
+
+client = AgentCampClient(project="support-agent", mode=Mode.OBSERVE)
+graph = client.wrap_langgraph(graph)
+```
+
+## Modes
+
+| Mode       | What it does                                                        |
+| ---------- | ------------------------------------------------------------------- |
+| `OBSERVE`  | Collect traces from the teacher agent only.                         |
+| `TRAIN`    | Build a fine-tuned expert from collected traces.                    |
+| `SHADOW`   | Run the expert alongside the teacher without serving it.            |
+| `ACTIVATE` | Serve the expert, falling back to the teacher when needed.          |
+
+## Configuration
+
+| Environment variable | Purpose                                  |
+| -------------------- | ---------------------------------------- |
+| `AGENTCAMP_API_KEY`  | API key used to authenticate the client. |
+| `AGENTCAMP_BASE_URL` | Override the AgentCamp API base URL.     |
+
+## Development
+
+```bash
+python -m venv .venv
+source .venv/bin/activate
+pip install -e '.[dev]'
+pytest
+```
+
+## License
+
+MIT — see [LICENSE](LICENSE).

agentcamp-0.1.0/src/agentcamp.egg-info/SOURCES.txt

@@ -0,0 +1,14 @@
+CHANGELOG.md
+LICENSE
+MANIFEST.in
+README.md
+pyproject.toml
+src/agentcamp/__init__.py
+src/agentcamp/client.py
+src/agentcamp/py.typed
+src/agentcamp.egg-info/PKG-INFO
+src/agentcamp.egg-info/SOURCES.txt
+src/agentcamp.egg-info/dependency_links.txt
+src/agentcamp.egg-info/requires.txt
+src/agentcamp.egg-info/top_level.txt
+tests/test_smoke.py

agentcamp-0.1.0/src/agentcamp.egg-info/dependency_links.txt

@@ -0,0 +1 @@
+

agentcamp-0.1.0/src/agentcamp.egg-info/requires.txt

@@ -0,0 +1,6 @@
+
+[dev]
+pytest
+pytest-cov
+build
+twine

agentcamp-0.1.0/src/agentcamp.egg-info/top_level.txt

@@ -0,0 +1 @@
+agentcamp

agentcamp-0.1.0/tests/test_smoke.py

@@ -0,0 +1,27 @@
+import agentcamp
+from agentcamp import AgentCampClient, AgentCampError, Mode, wrap_langgraph
+
+
+def test_version_exposed():
+    assert isinstance(agentcamp.__version__, str)
+    assert agentcamp.__version__.count(".") == 2
+
+
+def test_client_requires_project():
+    try:
+        AgentCampClient(project="")
+    except AgentCampError:
+        pass
+    else:
+        raise AssertionError("empty project should raise AgentCampError")
+
+
+def test_client_defaults():
+    client = AgentCampClient(project="support-agent")
+    assert client.project == "support-agent"
+    assert client.mode is Mode.OBSERVE
+
+
+def test_wrap_langgraph_is_passthrough():
+    sentinel = object()
+    assert wrap_langgraph(sentinel, project="support-agent") is sentinel