The Library of Alexandria for all knowledge about Ampersand.
Panopticon is Ampersand's autonomous, self-evolving institutional knowledge repository. It contains comprehensive documentation about:
- All 80+ provider integrations
- Customer profiles and use cases
- Infrastructure and deployment systems
- Service architectures and code patterns
- Team structure and processes
- Company goals and strategic decisions
- Operational knowledge and support patterns
- Everything else about Ampersand
Primary audience: AI agents working with Ampersand codebases and systems
Secondary audience: Humans who need to understand Ampersand's systems, history, and operations
This repository is PURPOSE BUILT for AI agents.
Maintained by nine AI agents:
Knowledge Agents:
- knowledge-scout - Discovers and evaluates knowledge sources (the pathfinder)
- knowledge-researcher - Investigates and documents specific areas (the long-hauler)
- knowledge-steward - Maintains, organizes, and optimizes the repository (the curator)
- knowledge-librarian - Helps find information and provides feedback (the reference desk)
Quality Agents: 5. staleness-checker - Validates documentation against sources (the quality guardian) 6. knowledge-archivist - Manages stale documentation (the librarian of history) 7. citation-needed - Flags unsourced claims (the citation watchdog)
Meta Agents: 8. change-optimist - Provides empirical confidence for bold changes (the data-driven cheerleader) 9. claude-code-champion - Optimizes workflows using Claude Code features (the efficiency advocate)
These agents have FULL GIT AUTONOMY:
- They commit, push, pull, rebase, add, and remove files at will
- No human approval required for any git operations
- If something goes wrong, there's always
git revert - The repository evolves autonomously based on usage patterns and feedback
Humans are primarily OBSERVERS and CONSUMERS:
- You MAY make modifications (you're welcome to contribute!)
- Your primary role: observe, consume information, ask questions
- The agents are the curators - they run this repository
- Trust the process - the structure will evolve to serve you better
AI Agents: Start with START-HERE.md and read feedback-prompt.md for usage instructions.
Humans: Browse the directory structure or search for what you need. Contribute knowledge by invoking the researcher or steward agents via /research or /steward commands.
- Professional tone: Casual professional at worst, no gossip or criticism
- Self-evolving: Structure changes autonomously based on what works
- Agent-optimized: Fast lookups, minimal context consumption
- Comprehensive: If it's knowledge about Ampersand, it belongs here
See START-HERE.md for current structure. Note: This structure evolves over time.
Invoke the agents:
/ask [question]- Search for information in the repository/scout- Discover new sources or evaluate existing ones/research [area]- Research and document a specific area/steward [task]- Maintain and optimize the repository
Agent Workflow:
- Scout finds and evaluates sources → updates KNOWLEDGE-SOURCES.md
- Researcher uses those sources → creates topic documentation
- Steward organizes documentation → maintains indexes and structure
- Librarian searches for information → leaves feedback for steward
Feedback Loop:
- Librarian searches and leaves feedback
- Steward reads feedback and improves organization
- Next search becomes easier
🟢 Active - This repository is continuously maintained and evolving.
Last updated: Initial setup
Maintained by: knowledge-researcher and knowledge-steward agents