Architecture Brain Tests — AI skill for Claude Code
End-to-end tests for AI agent Telegram gateway and memory management.
How to install Architecture Brain Tests
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open qwwiwi/architecture-brain-tests and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Architecture Brain Tests does
End-to-end tests for AI agent Telegram gateway and memory management.
Alternatives in AI
- Axonhub — ⚡️ Open-source AI Gateway — Use any SDK to call 100+ LLMs 3.2k ★
- Iai Personal Memory Engine — A cyber brain for your AI 866 ★
- Obsidian + Claude Code: Rebuild Your Second Brain With AI — Build an AI-powered personal knowledge management system from scratch 540 ★
README
Architecture Brain Tests
End-to-end and unit tests for AI agent Telegram gateway and memory management system.
Covers the full pipeline: **Telegram message -> Gateway -> Claude Code -> Response -> Memory**.
What This Tests
| Group | What | Type |
|---|---|---|
| T1: Config | Gateway config loading, validation, defaults | Unit |
| T2: Message Parsing | Telegram update -> internal format, allowlist, group gating | Unit |
| T3: OOB Commands | /stop, /reset, /status routing | Unit |
| T4: Media | File download, audio transcription (mock Groq) | Unit |
| T5: Session | New session creation, resume existing | Unit |
| T6: Claude Invoke | Subprocess args, workspace, model, permissions | Integration |
| T7: Response Format | Markdown -> HTML, tables, code blocks, split >4096 chars | Unit |
| T8: HOT Memory | append_to_hot_memory, fcntl lock, emergency trim | Unit |
| T9: OpenViking | push_to_openviking, filtering (own_text yes, external skip) | Unit |
| T10: Compression | trim-hot.sh input/output, bash fallback, Sonnet mock | Integration |
| T11: WARM Rotation | rotate-warm.sh 14d cutoff, move to COLD | Integration |
| T12: End-to-End | Full pipeline: Telegram -> Gateway -> Claude -> Memory | E2E |
| T13: Agent Laws | AGENT-LAWS.md structure, 9 principles, superpowers | Unit |
| T14: Structure | Repo structure, templates, skills, install.sh, Vibe Kanban | Unit |
| T15: Content API | UUID detection, video_id extraction, tariff access | Unit |
| T16: Skills | 11 base skills, SKILL.md frontmatter, no secrets | Unit |
| T17: Settings | settings.json template, permissions, hooks format, MCP | Unit |
| T18: Hooks | Lifecycle events, handler types, enforcement, examples | Unit |
| T19: Cron Pipeline | Script safety, flock, Sonnet fallback, cron order | Integration |
| T20: Security | No secrets in templates, secrets paths, git safety | Unit |
| T21: Context Budget | Token budgets, |
Related Skills
Catclaw
Personal AI assistant gateway powered by Claude Code — multi-agent, multi-channel (Discord/Telegram/TUI), sess
Opendray
Self-hosted gateway for Claude Code, Codex, Antigravity, Grok Build, OpenCode. Run AI coding agents on your ow
Openclaw Cookbook
Production-ready recipes for OpenClaw — self-hosted multi-channel AI agent gateway. Telegram, Discord, Slack,
Soulcast
Soulcast — Open-source AI agent gateway for Telegram. Built with Bun + TypeScript. Bot identity, persistent me
Symaira Brain
🧠🔐 Portable agent-context layer — multiplexes vault, memory, and skills behind one MCP gateway for AI coding
Brain CLI
Persistent memory for AI agents and personal knowledge management. Single-file Python CLI: hybrid semantic + k
Related Agents
Brain Agent
OneCommand's intelligence orchestrator. Runs at build start (READ memories, detect similar projects), after ea
Part 13: Memory Bridge (Give Coding Agents Your Brain)
When you spawn Codex or Claude Code to build something, they start blind. They don't know your architecture de
Coder Systems
Implements features for systems programming languages (C, C++, Rust, Go, Zig) to make tests pass. Expert in lo