- Add error handling for git fetch in Phase 0
- Add error handling for pnpm install and build in Phase 1
- Fix heredoc quoting to allow variable expansion in recovery prompt
- Use time-based health check loop instead of counter-based
- Standardize changelog path to ~/.clawdbot/molt/changelog.md
- Make wake command syntax consistent (--mode now)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Add detailed cron job prompts for nightly update and morning report
- Add notify.target field to config.json (use platform user IDs, not usernames)
- Add troubleshooting section for common notification issues
- Document freshness check (30h window) to avoid stale reports
- Explain message tool fallback behavior
After a successful rollback, molt now triggers the clawdbot agent to
diagnose and fix the issue. The agent receives crash logs and context,
attempts common fixes, and reports to the user if manual intervention
is needed.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Molt is a self-healing update system for self-hosted Clawdbot instances.
It handles nightly updates from upstream with automatic rollback if
something breaks.
Key features:
- Pulls from upstream, installs deps, builds, restarts gateway
- Health check with stability window (catches crash loops)
- Automatic rollback to pre-update commit on failure
- Module manifest to define what *you* care about (personalized health checks)
- Changelog generation for morning reports
- Agentic recovery philosophy: capture context, let the AI fix edge cases
Files:
- docs/molt.md: Full PRD with architecture, config schema, and rationale
- scripts/molt.sh: Reference implementation (bash)
This is an RFC seeking feedback on approach and integration strategy.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
* feat: audit fixes and documentation improvements
- Refactored model selection to drop legacy fallback and add warning
- Improved heartbeat content validation
- Added Skill Creation guide
- Updated CONTRIBUTING.md with roadmap
* style: fix formatting in model-selection.ts
* style: fix formatting and improve model selection logic with tests
Add documentation for running Clawdbot in a sandboxed macOS VM
using Lume. This provides an alternative to buying dedicated
hardware or using cloud instances.
The guide covers:
- Installing Lume on Apple Silicon Macs
- Creating and configuring a macOS VM
- Installing Clawdbot inside the VM
- Running headlessly for 24/7 operation
- iMessage integration via BlueBubbles
- Saving golden images for easy reset
Venice AI is a privacy-focused AI inference provider with support for
uncensored models and access to major proprietary models via their
anonymized proxy.
This integration adds:
- Complete model catalog with 25 models:
- 15 private models (Llama, Qwen, DeepSeek, Venice Uncensored, etc.)
- 10 anonymized models (Claude, GPT-5.2, Gemini, Grok, Kimi, MiniMax)
- Auto-discovery from Venice API with fallback to static catalog
- VENICE_API_KEY environment variable support
- Interactive onboarding via 'venice-api-key' auth choice
- Model selection prompt showing all available Venice models
- Provider auto-registration when API key is detected
- Comprehensive documentation covering:
- Privacy modes (private vs anonymized)
- All 25 models with context windows and features
- Streaming, function calling, and vision support
- Model selection recommendations
Privacy modes:
- Private: Fully private, no logging (open-source models)
- Anonymized: Proxied through Venice (proprietary models)
Default model: venice/llama-3.3-70b (good balance of capability + privacy)
Venice API: https://api.venice.ai/api/v1 (OpenAI-compatible)
