v0.0.11 Patch Release

.github/ISSUE_TEMPLATE/spec.yml

@@ -0,0 +1,28 @@
+name: Spec
+description: Technical specification for a new module, feature, or system change
+title: "[Spec]: "
+body:
+  - type: textarea
+    id: spec
+    attributes:
+      label: Specification
+      description: Full technical spec in markdown
+      value: |
+        ## Summary
+        <!-- One sentence: what this adds/changes and why -->
+
+        ## Motivation
+        <!-- What problem does this solve? Link related issues. -->
+
+        ## Design
+
+        ### API / Interface
+        <!-- Public API, CLI flags, stream types, config options -->
+
+        ### Architecture
+        <!-- How it fits into DimOS. Which modules, streams, blueprints are involved? -->
+
+        ### Implementation Notes
+        <!-- Key decisions, algorithms, dependencies, constraints -->
+    validations:
+      required: true

.github/workflows/macos.yml

@@ -0,0 +1,125 @@
+name: macos
+
+on:
+  push:
+    branches:
+      - main
+      - dev
+    paths-ignore:
+      - '**.md'
+  pull_request:
+    types: [opened, synchronize, reopened, ready_for_review]
+    paths-ignore:
+      - '**.md'
+      - 'docker/**'
+      - '.github/workflows/docker.yml'
+      - '.github/workflows/_docker-build-template.yml'
+
+permissions:
+  contents: read
+
+jobs:
+  check-changes:
+    if: github.event_name != 'pull_request' || github.event.pull_request.draft == false
+    runs-on: ubuntu-latest
+    outputs:
+      should-run: ${{ steps.filter.outputs.python }}
+    steps:
+      - uses: actions/checkout@v4
+      - id: filter
+        uses: dorny/paths-filter@v3
+        with:
+          filters: |
+            python:
+              - 'dimos/**'
+              - 'pyproject.toml'
+              - 'uv.lock'
+              - '.github/workflows/macos.yml'
+
+  macos-tests:
+    needs: [check-changes]
+    if: needs.check-changes.outputs.should-run == 'true'
+    runs-on: [self-hosted, macos, arm64]
+    timeout-minutes: 60
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          lfs: false
+
+      - name: Install uv
+        uses: astral-sh/setup-uv@v6
+        with:
+          enable-cache: true
+
+      - name: Set up Python
+        run: uv python install 3.12
+
+      - name: Install system dependencies
+        run: brew install gnu-sed gcc portaudio git-lfs libjpeg-turbo
+
+      - name: Set up git-lfs
+        run: |
+          git lfs install
+          git lfs pull
+
+      - name: Install dependencies
+        run: |
+          uv sync --all-extras --no-extra dds --no-extra cuda --frozen
+
+      - name: Check disk usage
+        run: |
+          df -h .
+          du -sh .venv/ || true
+
+      - name: Configure LCM networking
+        run: |
+          # Same as dimos autoconf for macOS (skipped when CI=1)
+          sudo route add -net 224.0.0.0/4 -interface lo0 || true
+          sudo sysctl -w kern.ipc.maxsockbuf=6291456
+          sudo sysctl -w net.inet.udp.recvspace=2097152
+          sudo sysctl -w net.inet.udp.maxdgram=65535
+
+      - name: Run tests
+        env:
+          OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
+          ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
+          ALIBABA_API_KEY: ${{ secrets.ALIBABA_API_KEY }}
+          LCM_DEFAULT_URL: "udpm://239.255.76.67:7667?ttl=0"
+          CI: "1"
+        run: |
+          source .venv/bin/activate
+          python -m pytest --durations=10 -m 'not (tool or mujoco)' --timeout=120 dimos/
+
+      - name: Check disk usage (post-test)
+        if: always()
+        run: df -h .
+
+  macos-mypy:
+    needs: [check-changes]
+    if: needs.check-changes.outputs.should-run == 'true'
+    runs-on: [self-hosted, macos, arm64]
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          lfs: false
+
+      - name: Install uv
+        uses: astral-sh/setup-uv@v6
+        with:
+          enable-cache: true
+
+      - name: Set up Python
+        run: uv python install 3.12
+
+      - name: Install system dependencies
+        run: brew install gnu-sed gcc portaudio git-lfs libjpeg-turbo
+
+      - name: Install dependencies
+        run: |
+          uv sync --all-extras --no-extra dds --no-extra cuda --frozen
+
+      - name: Run mypy
+        run: |
+          source .venv/bin/activate
+          mypy dimos
+# macOS CI retrigger

.github/workflows/stale.yml

@@ -0,0 +1,30 @@
+name: Close stale PRs
+
+on:
+  schedule:
+    - cron: "0 0 * * *"
+  workflow_dispatch:
+
+permissions:
+  pull-requests: write
+
+jobs:
+  stale:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/stale@v9
+        with:
+          days-before-stale: 23
+          days-before-close: 7
+          stale-pr-message: >
+            This pull request has been automatically marked as stale because it
+            has not had recent activity. It will be closed in 7 days if no
+            further activity occurs.
+          close-pr-message: >
+            This pull request has been automatically closed because it has been
+            stale for 30 days with no activity. Feel free to reopen it if you
+            plan to continue working on it.
+          stale-pr-label: stale
+          exempt-pr-labels: pinned,do-not-close
+          days-before-issue-stale: -1
+          days-before-issue-close: -1

.github/workflows/tests.yml

@@ -21,6 +21,7 @@ permissions:
 
 jobs:
   run-tests:
+    timeout-minutes: 60
     runs-on: [self-hosted, Linux]
     container:
       image: ghcr.io/dimensionalos/${{ inputs.dev-image }}

AGENTS.md

@@ -18,8 +18,7 @@ dimos list
 # --- Go2 quadruped ---
 dimos --replay run unitree-go2                  # perception + mapping, replay data
 dimos --replay run unitree-go2 --daemon         # same, backgrounded
-dimos --replay run unitree-go2-agentic          # + LLM agent (GPT-4o) + skills
-dimos --replay run unitree-go2-agentic-mcp      # + McpServer + McpClient (MCP tools live)
+dimos --replay run unitree-go2-agentic          # + LLM agent (GPT-4o) + skills + MCP server
 dimos run unitree-go2-agentic --robot-ip 192.168.123.161  # real Go2 hardware
 
 # --- G1 humanoid ---
@@ -39,11 +38,11 @@ dimos restart          # stop + re-run with same original args
 
 | Blueprint | Robot | Hardware | Agent | MCP server | Notes |
 |-----------|-------|----------|-------|------------|-------|
-| `unitree-go2-agentic-mcp` | Go2 | real | via McpClient | ✓ | **Only blueprint with McpServer live** |
+| `unitree-go2-agentic` | Go2 | real | via McpClient | ✓ | McpServer live |
 | `unitree-g1-agentic-sim` | G1 | sim | GPT-4o (G1 prompt) | — | Full agentic sim, no real robot needed |
 | `xarm-perception-agent` | xArm | real | GPT-4o | — | Manipulation + perception + agent |
 | `xarm7-trajectory-sim` | xArm7 | sim | — | — | Trajectory planning sim |
-| `arm-teleop-xarm7` | xArm7 | real | — | — | Quest VR teleop |
+| `teleop-quest-xarm7` | xArm7 | real | — | — | Quest VR teleop |
 | `dual-xarm6-planner` | xArm6×2 | real | — | — | Dual-arm motion planner |
 
 Run `dimos list` for the full list.
@@ -52,11 +51,11 @@ Run `dimos list` for the full list.
 
 ## Tools available to you (MCP)
 
-**MCP only works if the blueprint includes `McpServer`.** Currently the only shipped blueprint that does is `unitree-go2-agentic-mcp`. All other agentic blueprints use the in-process `Agent` module and do NOT expose an MCP endpoint.
+**MCP only works if the blueprint includes `McpServer`.** All shipped agentic blueprints use `McpServer` + `McpClient`. E.g.: `unitree-go2-agentic`.
 
 ```bash
 # Start the MCP-enabled blueprint first:
-dimos --replay run unitree-go2-agentic-mcp --daemon
+dimos --replay run unitree-go2-agentic --daemon
 
 # Then use MCP tools:
 dimos mcp list-tools                                              # all available skills as JSON
@@ -73,21 +72,21 @@ The MCP server runs at `http://localhost:9990/mcp` (`GlobalConfig.mcp_port`).
 
 ### Adding McpServer to a blueprint
 
-Use **both** `McpServer` and `mcp_client()` — do not mix with `agent()`.
+Use **both** `McpServer.blueprint()` and `McpClient.blueprint()`.
 
 ```python
-from dimos.agents.mcp.mcp_client import mcp_client
+from dimos.agents.mcp.mcp_client import McpClient
 from dimos.agents.mcp.mcp_server import McpServer
 
-unitree_go2_agentic_mcp = autoconnect(
+unitree_go2_agentic = autoconnect(
     unitree_go2_spatial,   # robot stack
     McpServer.blueprint(), # HTTP MCP server — exposes all @skill methods on port 9990
-    mcp_client(),          # LLM agent — fetches tools from McpServer
+    McpClient.blueprint(), # LLM agent — fetches tools from McpServer
     _common_agentic,       # skill containers
 )
 ```
 
-Reference: `dimos/robot/unitree/go2/blueprints/agentic/unitree_go2_agentic_mcp.py`
+Reference: `dimos/robot/unitree/go2/blueprints/agentic/unitree_go2_agentic.py`
 
 ---
 
@@ -276,7 +275,7 @@ my_skill_container = MySkillContainer.blueprint
 | Go2 (default) | `dimos/agents/system_prompt.py` | `SYSTEM_PROMPT` |
 | G1 humanoid | `dimos/robot/unitree/g1/system_prompt.py` | `G1_SYSTEM_PROMPT` |
 
-Pass the robot-specific prompt: `agent(system_prompt=G1_SYSTEM_PROMPT)`. Agent defaults to Go2 — wrong prompt causes hallucinated skills.
+Pass the robot-specific prompt: `McpClient.blueprint(system_prompt=G1_SYSTEM_PROMPT)`. The default prompt is Go2-specific; using it on G1 causes hallucinated skills.
 
 ### RPC Wiring
 
@@ -304,8 +303,6 @@ class MySkillContainer(Module):
 
 If multiple modules match the spec, use `.remappings()` to resolve. Source: `dimos/spec/utils.py`, `dimos/core/blueprints.py`.
 
-**Legacy**: existing skill containers use `rpc_calls: list[str]` + `get_rpc_calls("ClassName.method")`. This still works but wiring failures are silent and only surface at runtime. Don't use it in new code.
-
 ### Adding a New Skill
 
 1. Pick the right container (robot-specific or `dimos/agents/skills/`).

README.md

@@ -189,7 +189,7 @@ dimos run unitree-go2
 |-------------|-------------|
 | `dimos --replay run unitree-go2` | Quadruped navigation replay — SLAM, costmap, A* planning |
 | `dimos --replay --replay-dir unitree_go2_office_walk2 run unitree-go2-temporal-memory` | Quadruped temporal memory replay |
-| `dimos --simulation run unitree-go2-agentic-mcp` | Quadruped agentic + MCP server in simulation |
+| `dimos --simulation run unitree-go2-agentic` | Quadruped agentic + MCP server in simulation |
 | `dimos --simulation run unitree-g1` | Humanoid in MuJoCo simulation |
 | `dimos --replay run drone-basic` | Drone video + telemetry replay |
 | `dimos --replay run drone-agentic` | Drone + LLM agent with flight skills (replay) |
@@ -204,7 +204,7 @@ dimos run unitree-go2
 The `dimos` CLI manages the full lifecycle — run blueprints, inspect state, interact with agents, and call skills via MCP.
 
 ```bash
-dimos run unitree-go2-agentic-mcp --daemon   # Start in background
+dimos run unitree-go2-agentic --daemon   # Start in background
 dimos status                              # Check what's running
 dimos log -f                              # Follow logs
 dimos agent-send "explore the room"       # Send agent a command
@@ -270,17 +270,19 @@ Blueprints are instructions for how to construct and wire modules. We compose th
 
 Blueprints can be composed, remapped, and have transports overridden if `autoconnect()` fails due to conflicting variable names or `In[]` and `Out[]` message types.
 
-A blueprint example that connects the image stream from a robot to an LLM Agent for reasoning and action execution.
+A blueprint example that connects the image stream from a robot to an MCP-backed LLM agent for reasoning and action execution.
 ```py
 from dimos.core.blueprints import autoconnect
 from dimos.core.transport import LCMTransport
 from dimos.msgs.sensor_msgs import Image
 from dimos.robot.unitree.go2.connection import go2_connection
-from dimos.agents.agent import agent
+from dimos.agents.mcp.mcp_client import McpClient
+from dimos.agents.mcp.mcp_server import McpServer
 
 blueprint = autoconnect(
     go2_connection(),
-    agent(),
+    McpServer.blueprint(),
+    McpClient.blueprint(),
 ).transports({("color_image", Image): LCMTransport("/color_image", Image)})
 
 # Run the blueprint