mcp-as-code 0.1.4__tar.gz → 0.1.6__tar.gz

{mcp_as_code-0.1.4 → mcp_as_code-0.1.6}/PKG-INFO

@@ -1,10 +1,12 @@
 Metadata-Version: 2.4
 Name: mcp-as-code
-Version: 0.1.4
+Version: 0.1.6
 Summary: Execute MCP tools through generated Python code interfaces
 License-Expression: Apache-2.0
 License-File: LICENSE
 Requires-Python: >=3.11
+Requires-Dist: click>=8.1
+Requires-Dist: httpx>=0.28
 Requires-Dist: jinja2>=3.1
 Requires-Dist: matchlock==0.2.15
 Requires-Dist: mcp>=1.24.0
@@ -15,6 +17,8 @@ Description-Content-Type: text/markdown
 
 **Connect every MCP server you need, keeping your agent's context lean.**
 
+https://github.com/user-attachments/assets/4b91ea97-d48e-41c5-8189-0da8522ac459
+
 As the number of MCP servers you connect grows, tool schemas and intermediate tool call results clutter your agent's context. `maco` (mcp-as-code) collapses them all into a single endpoint with a programmatic interface.
 
 Instead of loading hundreds if not thousands of tool schemas upfront, `maco` reconstructs every MCP tool as Pydantic models and Python functions in a virtual filesystem and hands your agent just two of its favourite tools: `bash` to navigate, and `code_execute` to run. The agent discovers and composes tools as code, the thing frontier models do best.
@@ -40,26 +44,30 @@ tools
 **Programmatic leverage:** the agent is given a real programming language, Python, allowing it to orchestrate complex control flows with exceptional context-efficiency using loops, conditions, and state management.
 
 ```python
+import asyncio
 from collections import Counter
-from tools.github import listCommits
-
-owner, repo, page, counts = "openclaw", "openclaw", 1, Counter()
-
-while True:
-    commits = listCommits(owner=owner, repo=repo, perPage=100, page=page)
-    for commit in commits:
-        login = (commit.get("author") or {}).get("login")
-        if login and "bot" not in login.lower():
-            counts[login] += 1
-    if len(commits) < 100 or page >= 20:
-        break
-    page += 1
-
-total = sum(counts.values())
-for login, count in counts.most_common():
-    if count / total < 0.01:
-        break
-    print(f"@{login}: {count} commits ({count / total:.1%})")
+from tools.github import ListCommitsInput, list_commits
+
+async def main():
+    owner, repo, page, counts = "openclaw", "openclaw", 1, Counter()
+
+    while True:
+        commits = await list_commits(ListCommitsInput(owner=owner, repo=repo, per_page=100, page=page))
+        for commit in commits:
+            login = (commit.get("author") or {}).get("login")
+            if login and "bot" not in login.lower():
+                counts[login] += 1
+        if len(commits) < 100 or page >= 20:
+            break
+        page += 1
+
+    total = sum(counts.values())
+    for login, count in counts.most_common():
+        if count / total < 0.01:
+            break
+        print(f"@{login}: {count} commits ({count / total:.1%})")
+
+asyncio.run(main())
 ```
 
 The example above illustrates the MCP code that will be executed to find the top contributors to an open-source repository.
@@ -102,7 +110,7 @@ This config needs `npx` (for Playwright MCP), a GitHub token in `GITHUB_TOKEN`,
 Start the `maco` MCP server:
 
 ```bash
-maco serve-mcp --config mcp.json --provider docker
+maco up --config mcp.json --provider docker
 ```
 
 Use `--provider local` for a faster, non-isolated local feedback loop.
@@ -139,9 +147,16 @@ See [`docs/mcp-config.md`](docs/mcp-config.md) for the full config reference, in
 
 Choose the execution provider with `--provider`:
 
-- `local`: fastest feedback loop; runs commands as local subprocesses.
-- `docker`: runs commands in a long-lived Docker container.
-- `matchlock`: runs commands in a long-lived Matchlock micro-VM.
+- `local`: ideally for local development and fast feedback loop, or maco is already running in an isolated sandbox.
+- `docker`: runs mcp bash and code execution in a long-lived Docker container.
+- `matchlock`: runs mcp bash and code execution in a long-lived Matchlock micro-VM.
+
+## Credits
+
+`maco` is inspired by and builds on ideas from:
+
+- [Code execution with MCP: Building more efficient agents](https://www.anthropic.com/engineering/code-execution-with-mcp)
+- [Code Mode: the better way to use MCP](https://blog.cloudflare.com/code-mode/)
 
 ## License
 

{mcp_as_code-0.1.4 → mcp_as_code-0.1.6}/README.md

@@ -2,6 +2,8 @@
 
 **Connect every MCP server you need, keeping your agent's context lean.**
 
+https://github.com/user-attachments/assets/4b91ea97-d48e-41c5-8189-0da8522ac459
+
 As the number of MCP servers you connect grows, tool schemas and intermediate tool call results clutter your agent's context. `maco` (mcp-as-code) collapses them all into a single endpoint with a programmatic interface.
 
 Instead of loading hundreds if not thousands of tool schemas upfront, `maco` reconstructs every MCP tool as Pydantic models and Python functions in a virtual filesystem and hands your agent just two of its favourite tools: `bash` to navigate, and `code_execute` to run. The agent discovers and composes tools as code, the thing frontier models do best.
@@ -27,26 +29,30 @@ tools
 **Programmatic leverage:** the agent is given a real programming language, Python, allowing it to orchestrate complex control flows with exceptional context-efficiency using loops, conditions, and state management.
 
 ```python
+import asyncio
 from collections import Counter
-from tools.github import listCommits
-
-owner, repo, page, counts = "openclaw", "openclaw", 1, Counter()
-
-while True:
-    commits = listCommits(owner=owner, repo=repo, perPage=100, page=page)
-    for commit in commits:
-        login = (commit.get("author") or {}).get("login")
-        if login and "bot" not in login.lower():
-            counts[login] += 1
-    if len(commits) < 100 or page >= 20:
-        break
-    page += 1
-
-total = sum(counts.values())
-for login, count in counts.most_common():
-    if count / total < 0.01:
-        break
-    print(f"@{login}: {count} commits ({count / total:.1%})")
+from tools.github import ListCommitsInput, list_commits
+
+async def main():
+    owner, repo, page, counts = "openclaw", "openclaw", 1, Counter()
+
+    while True:
+        commits = await list_commits(ListCommitsInput(owner=owner, repo=repo, per_page=100, page=page))
+        for commit in commits:
+            login = (commit.get("author") or {}).get("login")
+            if login and "bot" not in login.lower():
+                counts[login] += 1
+        if len(commits) < 100 or page >= 20:
+            break
+        page += 1
+
+    total = sum(counts.values())
+    for login, count in counts.most_common():
+        if count / total < 0.01:
+            break
+        print(f"@{login}: {count} commits ({count / total:.1%})")
+
+asyncio.run(main())
 ```
 
 The example above illustrates the MCP code that will be executed to find the top contributors to an open-source repository.
@@ -89,7 +95,7 @@ This config needs `npx` (for Playwright MCP), a GitHub token in `GITHUB_TOKEN`,
 Start the `maco` MCP server:
 
 ```bash
-maco serve-mcp --config mcp.json --provider docker
+maco up --config mcp.json --provider docker
 ```
 
 Use `--provider local` for a faster, non-isolated local feedback loop.
@@ -126,9 +132,16 @@ See [`docs/mcp-config.md`](docs/mcp-config.md) for the full config reference, in
 
 Choose the execution provider with `--provider`:
 
-- `local`: fastest feedback loop; runs commands as local subprocesses.
-- `docker`: runs commands in a long-lived Docker container.
-- `matchlock`: runs commands in a long-lived Matchlock micro-VM.
+- `local`: ideally for local development and fast feedback loop, or maco is already running in an isolated sandbox.
+- `docker`: runs mcp bash and code execution in a long-lived Docker container.
+- `matchlock`: runs mcp bash and code execution in a long-lived Matchlock micro-VM.
+
+## Credits
+
+`maco` is inspired by and builds on ideas from:
+
+- [Code execution with MCP: Building more efficient agents](https://www.anthropic.com/engineering/code-execution-with-mcp)
+- [Code Mode: the better way to use MCP](https://blog.cloudflare.com/code-mode/)
 
 ## License
 

{mcp_as_code-0.1.4 → mcp_as_code-0.1.6}/docs/mcp-config.md

@@ -1,6 +1,6 @@
 # MCP configuration
 
-`maco` reads Claude-style MCP configuration files with a top-level `mcpServers` object. The same config is used by `maco gen`, `maco serve`, and `maco serve-mcp`.
+`maco` reads Claude-style MCP configuration files with a top-level `mcpServers` object. The same config is used by `maco gen` and `maco up`.
 
 ```json
 {

{mcp_as_code-0.1.4 → mcp_as_code-0.1.6}/examples/serve-mcp/README.md

@@ -1,6 +1,6 @@
-# `maco serve-mcp` example
+# `maco up` example
 
-This example shows how to expose several upstream MCP servers through one compact `maco serve-mcp` endpoint. The upstream servers here are:
+This example shows how to expose several upstream MCP servers through one compact `maco up` endpoint. The upstream servers here are:
 
 - [Playwright MCP](https://playwright.dev/mcp/introduction), launched with `npx -y @playwright/mcp@latest`
 - [GitHub MCP server](https://github.com/github/github-mcp-server), using GitHub's hosted Streamable HTTP endpoint at `https://api.githubcopilot.com/mcp/`
@@ -19,20 +19,20 @@ If you are already authenticated with the GitHub CLI, export a token directly:
 export GITHUB_TOKEN=$(gh auth token)
 ```
 
-## 1. Start `maco serve-mcp`
+## 1. Start `maco up`
 
 The short path is to run from this example directory so the defaults line up with the local files:
 
 ```bash
 cd examples/serve-mcp
-uv run maco serve-mcp --provider local
+uv run maco up --provider local
 ```
 
-This uses `mcp.json`, writes `.maco/gateway.json`, uses `maco-serve-mcp/` as scratch, starts the gateway, and serves HTTP MCP at `http://127.0.0.1:8789/mcp`. Add `--clean` only when you want to recreate the local generated SDK from scratch.
+This uses `mcp.json`, writes `.maco/gateway.json`, uses `.maco/scratch` as the default scratch directory, starts the gateway, and serves HTTP MCP at `http://127.0.0.1:8789/mcp`. Add `--clean` only when you want to recreate the local generated SDK from scratch.
 
 ## 2. Connect an agent to the MCP gateway
 
-Configure your MCP client with `mcp-client.json` to connect to `maco serve-mcp`. For example, in your MCP client's settings, set the MCP server URL to `http://127.0.0.1:8789/mcp`.
+Configure your MCP client with `mcp-client.json` to connect to `maco up`. For example, in your MCP client's settings, set the MCP server URL to `http://127.0.0.1:8789/mcp`.
 
 If you are using codex you may connect to it via:
 

{mcp_as_code-0.1.4 → mcp_as_code-0.1.6}/images/sandbox/README.md

@@ -1,20 +1,20 @@
 # maco sandbox image
 
-This directory contains the Docker image used by the Docker and Matchlock `maco serve-mcp` sandbox providers.
+This directory contains the Docker image used by the Docker and Matchlock `maco up` sandbox providers.
 
 The image extends the pinned Alpine uv/Python image with the `maco` CLI, `ripgrep` (`rg`), and `fd`. Remote sandbox providers use the CLI to bootstrap `/workspace/macosdk/tools` from the live gateway catalog at sandbox startup.
 
 ## Build
 
 ```bash
-VERSION="$(cat VERSION.txt)"
+VERSION="$(scripts/package-version)"
 docker build -f images/sandbox/Dockerfile -t "ghcr.io/jingkaihe/maco:${VERSION}-alpine" .
 ```
 
 ## Import into Matchlock
 
 ```bash
-VERSION="$(cat VERSION.txt)"
+VERSION="$(scripts/package-version)"
 docker save "ghcr.io/jingkaihe/maco:${VERSION}-alpine" \
   | matchlock image import "ghcr.io/jingkaihe/maco:${VERSION}-alpine"
 ```

{mcp_as_code-0.1.4 → mcp_as_code-0.1.6}/pyproject.toml

@@ -6,6 +6,8 @@ readme = "README.md"
 license = "Apache-2.0"
 requires-python = ">=3.11"
 dependencies = [
+    "click>=8.1",
+    "httpx>=0.28",
     "jinja2>=3.1",
     "matchlock==0.2.15",
     "mcp>=1.24.0",
@@ -27,8 +29,7 @@ dev = [
 ]
 
 [tool.hatch.version]
-path = "VERSION.txt"
-pattern = "(?P<version>.+)"
+path = "src/maco/__init__.py"
 
 [tool.hatch.build.targets.wheel]
 packages = ["src/maco"]
@@ -39,6 +40,5 @@ include = [
     "docs",
     "README.md",
     "LICENSE",
-    "VERSION.txt",
     "pyproject.toml",
 ]

{mcp_as_code-0.1.4 → mcp_as_code-0.1.6}/src/maco/__init__.py

@@ -1,3 +1,3 @@
 """maco Python interface generator and MCP code-execution server."""
 
-__version__ = "0.1.0"
+__version__ = "0.1.6"