mirror of
https://github.com/instavm/open-skills.git
synced 2026-11-18 18:03:45 +00:00
demo video
This commit is contained in:
+2
-2
@@ -40,10 +40,10 @@ container rm buildkit
|
||||
container build --tag os --file Dockerfile .
|
||||
|
||||
# Tag the newly built container
|
||||
container images tag os bandarlabs/open-skills
|
||||
container image tag os instavm/open-skills
|
||||
|
||||
# Push the image to the registry
|
||||
container images push bandarlabs/open-skills
|
||||
container image push instavm/open-skills
|
||||
```
|
||||
|
||||
</details>
|
||||
|
||||
@@ -187,7 +187,7 @@
|
||||
same "printed page" as the copyright notice for easier
|
||||
identification within third-party archives.
|
||||
|
||||
Copyright 2025 Coderunner
|
||||
Copyright 2025 Open-Skills
|
||||
|
||||
Licensed under the Apache License, Version 2.0 (the "License");
|
||||
you may not use this file except in compliance with the License.
|
||||
|
||||
@@ -6,15 +6,24 @@
|
||||
|
||||
# Run Claude Skills Locally on Your Mac (No Cloud Upload Required)
|
||||
|
||||
Anthropic recently announced [Skills for Claude](https://www.anthropic.com/news/skills) - reusable folders with instructions, scripts, and resources that make Claude better at specialized tasks. This MCP server lets you run these skills **entirely on your local machine** in a sandboxed environment.
|
||||
Anthropic recently announced [Skills for Claude](https://www.anthropic.com/news/skills) - reusable folders with instructions, scripts, and resources that make Claude better at specialized tasks. This tool lets you run these skills **entirely on your local machine** in a sandboxed environment.
|
||||
|
||||
**What this means:** Claude can now process your files (documents, spreadsheets, presentations, images) using these specialized skills while keeping all data on your Mac. No uploads, complete privacy.
|
||||
**What this means:** You can now process your files (documents, spreadsheets, presentations, images) using these specialized skills while keeping all data on your Mac. No uploads, complete privacy.
|
||||
|
||||
> This MCP server executes AI-generated code in a truly isolated sandboxed environment on your Mac using Apple's native containers.
|
||||
> This tool executes AI-generated code in a truly isolated sandboxed environment on your Mac using Apple's native containers.
|
||||
|
||||
## Demo
|
||||
|
||||
Watch Open-Skills in action with Gemini CLI:
|
||||
|
||||
<video src="demo-gemini.mp4" controls width="100%">
|
||||
<source src="demo-gemini.mp4" type="video/mp4">
|
||||
Your browser does not support the video tag. <a href="demo-gemini.mp4">Download the demo video</a>.
|
||||
</video>
|
||||
|
||||
## Why Run Skills Locally?
|
||||
|
||||
- **Privacy:** Process sensitive documents, financial data, or proprietary files without cloud upload
|
||||
- **Privacy:** Process sensitive documents, financial data
|
||||
- **Full Control:** Skills execute in an isolated container with VM-level isolation
|
||||
- **Compatibility:** Works with Claude Desktop, Gemini CLI, Qwen CLI, or any MCP-compatible tool
|
||||
- **Extensibility:** Import Anthropic's official skills or create your own custom skills
|
||||
@@ -83,7 +92,7 @@ Edit `~/.gemini/settings.json`:
|
||||
"selectedAuthType": "oauth-personal",
|
||||
"mcpServers": {
|
||||
"open-skills": {
|
||||
"httpUrl": "http://coderunner.local:8222/mcp"
|
||||
"httpUrl": "http://open-skills.local:8222/mcp"
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
Binary file not shown.
@@ -10,7 +10,7 @@ def resolve_with_system_dns(hostname):
|
||||
print(f"Error resolving {hostname}: {e}")
|
||||
return None
|
||||
|
||||
hostname = "coderunner.local"
|
||||
hostname = "open-skills.local"
|
||||
address = resolve_with_system_dns(hostname)
|
||||
# Create a proxy directly from a config dictionary
|
||||
config = {
|
||||
|
||||
@@ -1,8 +1,8 @@
|
||||
We are currently on macbook, and whenever required we use tool to execute codes (in a jupyter like server). the code is executed in a container (you wouldn't notice but just know this).
|
||||
|
||||
The paths on local machine is ~/.coderunner/assets/skills/user is mapped to /app/uploads/skills/user inside container.
|
||||
The paths on local machine is ~/.open-skills/assets/skills/user is mapped to /app/uploads/skills/user inside container.
|
||||
|
||||
~/.coderunner/assets/outputs (in the host machine) is mapped to /app/uploads/outputs inside conatiner. This is where user will puts their files they want to edit like some png, pdf, txt etc. You should also use it to output your artifacts generated.
|
||||
~/.open-skills/assets/outputs (in the host machine) is mapped to /app/uploads/outputs inside conatiner. This is where user will puts their files they want to edit like some png, pdf, txt etc. You should also use it to output your artifacts generated.
|
||||
|
||||
So that will help whenever we need a file inside a container to work on it via the execute code tool.
|
||||
|
||||
|
||||
@@ -35,7 +35,7 @@ def resolve_with_system_dns(hostname):
|
||||
|
||||
|
||||
async def main():
|
||||
hostname = "coderunner.local"
|
||||
hostname = "open-skills.local"
|
||||
address = resolve_with_system_dns(hostname)
|
||||
async with MCPServerStreamableHttp(
|
||||
name="SSE Python Server",
|
||||
|
||||
+4
-4
@@ -63,8 +63,8 @@ echo "Starting the Sandbox Container..."
|
||||
container system start
|
||||
|
||||
|
||||
echo "Pulling the latest image: bandarlabs/open-skills"
|
||||
container image pull bandarlabs/open-skills
|
||||
echo "Pulling the latest image: instavm/open-skills"
|
||||
container image pull instavm/open-skills
|
||||
|
||||
echo "→ Ensuring open-skills assets directories…"
|
||||
ASSETS_SRC="$HOME/.open-skills/assets"
|
||||
@@ -72,7 +72,7 @@ mkdir -p "$ASSETS_SRC/skills/user"
|
||||
mkdir -p "$ASSETS_SRC/outputs"
|
||||
|
||||
# Run the command to start the sandbox container
|
||||
echo "Running: container run --name open-skills --detach --rm --cpus 8 --memory 4g bandarlabs/open-skills"
|
||||
echo "Running: container run --name open-skills --detach --rm --cpus 8 --memory 4g instavm/open-skills"
|
||||
container run \
|
||||
--volume "$ASSETS_SRC/skills/user:/app/uploads/skills/user" \
|
||||
--volume "$ASSETS_SRC/outputs:/app/uploads/outputs" \
|
||||
@@ -81,6 +81,6 @@ container run \
|
||||
--rm \
|
||||
--cpus 8 \
|
||||
--memory 4g \
|
||||
bandarlabs/open-skills
|
||||
instavm/open-skills
|
||||
|
||||
echo "✅ Setup complete. MCP server is available at http://open-skills.local:8222/mcp"
|
||||
Executable
+42
@@ -0,0 +1,42 @@
|
||||
#!/bin/bash
|
||||
|
||||
# Script to push container image with retry logic until success
|
||||
# Usage: ./push_with_retry.sh
|
||||
|
||||
IMAGE_NAME="instavm/open-skills:latest"
|
||||
MAX_ATTEMPTS=100
|
||||
ATTEMPT=1
|
||||
INITIAL_WAIT=5
|
||||
MAX_WAIT=60
|
||||
|
||||
echo "Starting container image push with retry logic..."
|
||||
echo "Image: $IMAGE_NAME"
|
||||
echo "Will retry until successful"
|
||||
echo ""
|
||||
|
||||
while true; do
|
||||
echo "----------------------------------------"
|
||||
echo "Attempt #$ATTEMPT ($(date '+%Y-%m-%d %H:%M:%S'))"
|
||||
echo "----------------------------------------"
|
||||
|
||||
if container image push "$IMAGE_NAME"; then
|
||||
echo ""
|
||||
echo "✓ SUCCESS! Image pushed successfully on attempt #$ATTEMPT"
|
||||
exit 0
|
||||
else
|
||||
EXIT_CODE=$?
|
||||
echo ""
|
||||
echo "✗ Push failed with exit code $EXIT_CODE"
|
||||
|
||||
# Calculate wait time with exponential backoff
|
||||
WAIT_TIME=$((INITIAL_WAIT * (ATTEMPT > 3 ? 2 : 1)))
|
||||
if [ $WAIT_TIME -gt $MAX_WAIT ]; then
|
||||
WAIT_TIME=$MAX_WAIT
|
||||
fi
|
||||
|
||||
echo "Waiting ${WAIT_TIME} seconds before retry..."
|
||||
sleep $WAIT_TIME
|
||||
|
||||
ATTEMPT=$((ATTEMPT + 1))
|
||||
fi
|
||||
done
|
||||
@@ -29,7 +29,7 @@ logging.basicConfig(
|
||||
logger = logging.getLogger(__name__)
|
||||
|
||||
# Initialize the MCP server with a descriptive name for the toolset
|
||||
mcp = FastMCP("CodeRunner")
|
||||
mcp = FastMCP("Open-Skills")
|
||||
|
||||
# Kernel pool configuration
|
||||
MAX_KERNELS = 5
|
||||
@@ -567,7 +567,7 @@ async def _parse_skill_frontmatter(skill_md_path):
|
||||
@mcp.tool()
|
||||
async def list_skills() -> str:
|
||||
"""
|
||||
Lists all available skills in the CodeRunner container.
|
||||
Lists all available skills in the Open-Skills container.
|
||||
|
||||
Returns a list of available skills organized by category (public/user).
|
||||
Public skills are built into the container, while user skills are added by users.
|
||||
|
||||
Reference in New Issue
Block a user