PR #45 added fetch and sequentialthinking MCP sidecars to values.yaml
and the deployment template but missed updating values.schema.json.
The schema has additionalProperties: false on mcp.sidecars, causing
Helm upgrade to fail with validation errors.
Also adds resourceProfile to the schema as it was missing.
Generated with [Claude Code](https://claude.ai/code)
via [Happy](https://happy.engineering)
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: Happy <yesreply@happy.engineering>
Remove tag triggers and duplicate release job from build-and-push.yaml.
The release-unified.yaml workflow handles the full release flow (Docker
build, Helm chart, GitHub release) when triggered via workflow_dispatch.
Previously, release-unified.yaml pushing a commit to main AND a v* tag
would trigger build-and-push.yaml up to twice, causing multiple Docker
builds to race for the :latest tag. The stale GHA layer cache in the
racing build could overwrite :latest with an image missing new tools
(e.g., crush, opencode).
Changes:
- Remove tags: ['v*'] trigger (release-unified handles tag-based releases)
- Remove duplicate release job (Helm chart + GitHub release)
- Remove semver tag patterns from metadata (not needed without tag trigger)
- Skip builds from github-actions[bot] to avoid racing with release commits
Generated with [Claude Code](https://claude.ai/code)
via [Happy](https://happy.engineering)
Co-Authored-By: Claude <noreply@anthropic.com>
Co-Authored-By: Happy <yesreply@happy.engineering>
Add two new MCP (Model Context Protocol) sidecars to enable web content fetching
and structured problem-solving capabilities:
- **Fetch MCP**: Web content fetching and HTML to markdown conversion on port 8082
- **Sequential Thinking MCP**: Structured thinking and problem-solving processes on port 8083
Both sidecars are enabled by default and use the official MCP Docker images
(mcp/fetch and mcp/sequentialthinking) with fastmcp SSE transport.
Changes:
- Add fetch and sequentialthinking sidecars to values.yaml
- Add sidecar containers to deployment.yaml template
- Update .mcp.json with new server endpoints
- Update CLAUDE.md documentation with new sidecar details
Closes#43, #44
Generated with [Claude Code](https://claude.ai/code)
via [Happy](https://happy.engineering)
Co-Authored-By: Claude <noreply@anthropic.com>
Co-Authored-By: Happy <yesreply@happy.engineering>
- Force fresh APT repository data by clearing package cache before update
- Add debugging output to show available and installed versions
- Use --no-install-recommends to avoid unnecessary packages
- Add version validation during build process
- Disable auto-updates in Antigravity settings to prevent container conflicts
This should resolve the "agentSessions service not found" error that prevents
the AI chat window from responding.
Generated with [Claude Code](https://claude.ai/code)
via [Happy](https://happy.engineering)
Co-Authored-By: Claude <noreply@anthropic.com>
Co-Authored-By: Happy <yesreply@happy.engineering>
- Move release job into build-and-push workflow with dependency
- Remove separate release.yaml workflow to prevent race condition
- Ensures Docker image is available before GitHub release is published
Fixes the issue where clients see release with docker pull instructions
before the image is uploaded to GHCR.
Generated with [Claude Code](https://claude.ai/code)
via [Happy](https://happy.engineering)
Co-Authored-By: Claude <noreply@anthropic.com>
Co-Authored-By: Happy <yesreply@happy.engineering>
The Playwright MCP image (mcr.microsoft.com/playwright/mcp) doesn't
support --transport flag. Use node cli.js entrypoint with --headless,
--browser chromium, --no-sandbox, and --port for SSE mode.
Fixes#41
Generated with [Claude Code](https://claude.ai/code)
via [Happy](https://happy.engineering)
Co-Authored-By: Claude <noreply@anthropic.com>
Co-Authored-By: Happy <yesreply@happy.engineering>
- Fix CLAUDE.md: update Docker build example from antigravity to devcontainer
- Fix README.md stale value references throughout:
- ide → ide.type, ssh → ssh.enabled
- happyServerUrl → happy.serverUrl (and all happy.* values)
- secureConnection → display.secureConnection
- userId/groupId → user.id/user.groupId
- mcpSidecars.* → mcp.sidecars.*
- Remove GitHub MCP sidecar from custom config example (discontinued)
- Fix YAML indentation in MCP config examples
- Fix storage mount path from /home to /config
- Fix Happy state directory from /home/user/.happy to /config/userdata/.happy
Generated with [Claude Code](https://claude.ai/code)
via [Happy](https://happy.engineering)
Co-Authored-By: Claude <noreply@anthropic.com>
Co-Authored-By: Happy <yesreply@happy.engineering>
Three fixes for broken MCP sidecars:
1. Remove GitHub MCP sidecar entirely - the upstream image
(ghcr.io/modelcontextprotocol/servers/github) is discontinued.
GitHub MCP is now accessed via Copilot API instead.
2. Fix Playwright MCP image - changed from non-existent
microsoft/playwright-mcp to mcr.microsoft.com/playwright/mcp
(the official Microsoft Container Registry image).
3. Fix pgtuner MCP startup command - changed --transport to --mode
and added --host 0.0.0.0 to match the current pgtuner_mcp CLI.
Bumps chart version to 0.3.3.
Closes#40
Generated with [Claude Code](https://claude.ai/code)
via [Happy](https://happy.engineering)
Co-Authored-By: Claude <noreply@anthropic.com>
Co-Authored-By: Happy <yesreply@happy.engineering>
service.yaml was still using the old flat values format (.Values.ide
and .Values.ssh) while values.yaml had been restructured to nested
objects (ide.type and ssh.enabled). This caused a type comparison
error breaking all devcontainer deployments across cpfarhood and
homeassistant namespaces.
Bumps chart version to 0.3.1.
Generated with [Claude Code](https://claude.ai/code)
via [Happy](https://happy.engineering)
Co-Authored-By: Claude <noreply@anthropic.com>
Co-Authored-By: Happy <yesreply@happy.engineering>
Install both terminal-based AI coding agents alongside Happy Coder:
- OpenCode (opencode-ai/opencode) - open-source AI coding agent
- Crush (charmbracelet/crush) - OpenCode's active successor by Charm
Both are installed as Go binaries to /usr/local/bin from latest GitHub
releases and available to users via `opencode` and `crush` commands.
Generated with [Claude Code](https://claude.ai/code)
via [Happy](https://happy.engineering)
Co-Authored-By: Claude <noreply@anthropic.com>
Co-Authored-By: Happy <yesreply@happy.engineering>
Major documentation updates to reflect the reorganized Helm chart:
- Update README.md and CLAUDE.md to use new mcp.sidecars structure
- Reflect removal of Happy daemon automatic startup
- Document new logical values organization (Basic → Access → Infrastructure → Integrations → Smart Defaults)
- Update all MCP sidecar examples to use mcp.sidecars instead of mcpSidecars
- Document new quickstart deployment options and progressive disclosure
- Update troubleshooting sections with current configurations
- Standardize all secret environment variable references to SCREAMING_SNAKE_CASE
This completes the documentation sync following the major Helm chart
user experience improvements implemented in previous commits.
Generated with [Claude Code](https://claude.ai/code)
via [Happy](https://happy.engineering)
Co-Authored-By: Claude <noreply@anthropic.com>
Co-Authored-By: Happy <yesreply@happy.engineering>
Removes unreliable automatic Happy Coder daemon startup, allowing users
to start it manually when needed for better reliability and control.
**Changes:**
- scripts/init-repo.sh: Removed Happy daemon startup code and lock cleanup
- scripts/startapp.sh: Updated comment to reflect init-repo only handles git
- README.md: Updated startup flow documentation and troubleshooting section
- CLAUDE.md: Updated startup flow and file descriptions
**Benefits:**
- No more unreliable automatic daemon startup failures
- Users can start Happy daemon manually when needed: `happy daemon start`
- Cleaner container startup without Happy-related delays or errors
- Happy configuration and credentials still persist on PVC when used
**Usage:**
Users can now manually start Happy Coder when needed:
```bash
happy daemon start # Start when needed
happy daemon status # Check status
happy daemon stop # Stop if needed
```
Generated with [Claude Code](https://claude.ai/code)
via [Happy](https://happy.engineering)
Co-Authored-By: Claude <noreply@anthropic.com>
Co-Authored-By: Happy <yesreply@happy.engineering>
- Fix homeassistant MCP sidecar command by removing incorrect module specification
- Add init container for antigravity IDE to create /config/userdata directory
- Bump chart version to 0.2.2
The homeassistant sidecar was failing with "File not found: /app/ha_mcp.main"
because fastmcp should run without explicit module specification.
The antigravity IDE was failing to initialize due to missing /config/userdata
directory. Added an init container to ensure the directory exists with proper
permissions before the main container starts.
Generated with Claude Code via Happy
Co-Authored-By: Claude <noreply@anthropic.com>
Co-Authored-By: Happy <yesreply@happy.engineering>
The GitHub MCP server has been archived and moved to servers-archived repo.
There is no Docker image available at ghcr.io/modelcontextprotocol/servers/github.
Disabling by default to prevent ImagePullBackOff errors.
TODO: Either build custom image from archived source or find alternative.
Generated with [Claude Code](https://claude.ai/code)
via [Happy](https://happy.engineering)
Co-Authored-By: Claude <noreply@anthropic.com>
Co-Authored-By: Happy <yesreply@happy.engineering>
- Fixed Home Assistant MCP command flags from --sse-server-host/port to --host/port
- Added GitHub MCP server as new sidecar (enabled by default)
- Uses existing GITHUB_TOKEN from environment
- Updated documentation and .mcp.json configuration
The GitHub MCP sidecar provides AI assistants with ability to interact with
GitHub repositories, issues, PRs, and more using the same token used for
repository cloning.
Fixes#26
Generated with [Claude Code](https://claude.ai/code)
via [Happy](https://happy.engineering)
Co-Authored-By: Claude <noreply@anthropic.com>
Co-Authored-By: Happy <yesreply@happy.engineering>
Removed 4 redundant workflow files:
- helm-publish.yaml (integrated into release-unified)
- helm-release.yaml.old (the auto-version-bump nightmare)
- release.yaml (replaced by release-unified)
- tag-release.md (old docs, replaced by README.md)
Now down to just 3 clean workflows:
- release-unified.yaml: Complete release process
- quick-fix.yaml: Emergency Docker builds
- build-and-push.yaml: Automatic CI for PRs/tags
Also updated documentation to reflect the simplified pipeline.
Generated with [Claude Code](https://claude.ai/code)
via [Happy](https://happy.engineering)
Co-Authored-By: Claude <noreply@anthropic.com>
Co-Authored-By: Happy <yesreply@happy.engineering>
BREAKING CHANGE: Removed auto-version-bump workflow that was causing [skip ci] hell
New features:
- Unified release workflow that handles everything in one go
- Quick fix workflow for emergency deployments
- No more [skip ci] preventing Docker builds
- No more manual tag juggling
- Single button release process
The old pipeline was a disaster with disconnected workflows, auto-version-bumps
with [skip ci] that prevented Docker builds, and required manual tag deletion/
re-pushing to trigger builds. This new pipeline is actually usable.
Generated with [Claude Code](https://claude.ai/code)
via [Happy](https://happy.engineering)
Co-Authored-By: Claude <noreply@anthropic.com>
Co-Authored-By: Happy <yesreply@happy.engineering>
The Home Assistant MCP was failing with 'fastmcp-sse.json not found' error.
Updated command to use proper fastmcp arguments for SSE transport mode
without requiring a configuration file.
Generated with [Claude Code](https://claude.ai/code)
via [Happy](https://happy.engineering)
Co-Authored-By: Claude <noreply@anthropic.com>
Co-Authored-By: Happy <yesreply@happy.engineering>