- Create patterns/ directory with README, manifest, and 10 initial patterns
covering Ollama JSON fallback, API escaping, deprecation, PTY auth,
queue-poll, LLM-as-parser, credential rotation, reverse proxy binding,
human approval gates, and transient retry.
- Wire pattern loading into architecture/pipeline.js based on task tags.
- Update architecture/orchestrator.js to load patterns and surface them in
the system prompt.
- Update MEMORY.md, ARCHITECTURE.md, and CONTEXT.md to document the registry
and record the decision.
- Updated DREAMS.md and MEMORY.md with new entries
- Added daily dreaming session corpus (2026-07-05 through 2026-07-18)
- Added deep/light/REM dream analysis files for July 6-19
- Added Intune rollout outputs for KMCC client
- Added Vault API integration scripts and documentation
- Added container update playbook (update-containers.yml)
- Added .gitignore for generated/noise files
- Simplified to use direct shell loops for discovering running containers/VMs
- Uses zip filter pattern for pairing container IDs with OS types
- Fixes the bug where nested loop result dictionaries were passed
as container IDs instead of simple strings (causing container: not found)
- Matches the working pattern from update-containers.yml
- Set correct table ID: mx149yctebfwvys (ai_data_Memory)
- Updated type column with SingleSelect options
- Added all required columns for corrections, preferences, episodes, decisions, validation
- Fixed column options for type, severity, status fields
Adds infrastructure endpoints from vault:
- rustfs (S3-compatible) for object storage
- Qdrant for vector database
- NocoDB for structured data
- Metabase for analytics
- Odoo for business data
Documents how to:
- Authenticate with AppRole
- Read secrets from kv/data/api/
- Use service-specific roles for limited access
- Locate credential files
JC Beasley