2026-04-11 12:59:02 +02:00
|
|
|
# Glitch Unviersity Local Inference
|
2026-04-11 00:43:18 +02:00
|
|
|
|
2026-04-11 12:59:02 +02:00
|
|
|
TL;DR This repo is useful if you want to run Agent0 (probabily to use local inference) and also connect to Glitch Unviersity infrastructure.
|
2026-04-11 00:43:18 +02:00
|
|
|
|
|
|
|
|
## Tool Inventory
|
|
|
|
|
|
2026-04-11 12:59:02 +02:00
|
|
|
Agents operate inside an AgentZero container with the following tools:
|
2026-04-11 00:43:18 +02:00
|
|
|
|
|
|
|
|
### Core Workflow
|
|
|
|
|
- **gutask** — task management, orientation, session lifecycle, letters, lore, notes
|
|
|
|
|
- **git 2.51** — version control (commit, push, pull, branch, log, diff)
|
|
|
|
|
- **curl 8.18** — HTTP requests, API testing and debugging
|
|
|
|
|
|
|
|
|
|
### AgentZero Framework
|
|
|
|
|
- **code_execution_tool** — run terminal commands, Python, and Node.js
|
|
|
|
|
- **text_editor** — read, write, and patch files with line-level precision
|
|
|
|
|
- **browser_agent** — Playwright-based headless browser via subordinate agent
|
|
|
|
|
- **call_subordinate** — delegate tasks to specialized agents
|
|
|
|
|
- **document_query** — read and query remote/local documents
|
|
|
|
|
- **search_engine** — web search
|
|
|
|
|
- **memory tools** — long-term persistent memory
|
|
|
|
|
|
|
|
|
|
### Languages and Runtimes
|
|
|
|
|
- **Python 3.13** + pip + requests library
|
|
|
|
|
- **Node.js 22** + npm 9
|
|
|
|
|
|
|
|
|
|
### System Utilities
|
|
|
|
|
- wget, ssh, sed, awk, grep, apt (can install anything needed)
|
|
|
|
|
|
|
|
|
|
### Not Installed (Candidates)
|
|
|
|
|
- gh (GitHub CLI)
|
|
|
|
|
- jq (JSON processing)
|
|
|
|
|
|
|
|
|
|
# Guardrails
|
|
|
|
|
1. Always git fetch and pull main before starting work
|
|
|
|
|
1. Push to main branch, but create task that changes must be deployed and assign this deployment task to Glitch Hunter.
|
|
|
|
|
2. Never drop or truncate database tables or columns without explicit human sign-off.
|
|
|
|
|
3. Never run destructive operations (rm -rf, force push, hard reset) without confirming with Glitch Hunter first.
|
|
|
|
|
4. Never change the architecture — structure, patterns, tech choices — without Glitch Hunter's approval. Implement, don't redesign.
|
|
|
|
|
5. Never mark a task done unless it is verifiably working, not just theoretically complete.
|
|
|
|
|
6. Never store secrets, credentials, or keys in code or notes.
|
|
|
|
|
7. Never proceed on an ambiguous brief. Write a clarifying note and wait.
|
|
|
|
|
8. Never modify tasks or notes belonging to another agent without being asked.
|
