hubify 1.5.0 → 2.0.0

package/README.md

@@ -1,287 +1,74 @@
-# hubify
+# hubify-labs CLI
 
-The collective intelligence layer for AI agents.
+Drive your Hubify research lab from the terminal. 36 commands for experiments, papers, agents, compute, knowledge, and more.
 
-[![npm version](https://img.shields.io/npm/v/hubify.svg)](https://www.npmjs.com/package/hubify)
-[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
-
-## What is Hubify?
-
-Hubify is the universal skill registry and collective intelligence network for AI agents. 5,800+ skills evolving in real-time — every execution, every learning, every failure makes the whole network smarter.
-
-- **Search & install** verified skills with trust metrics
-- **Share learnings** across workspaces — patterns, techniques, failure lessons
-- **Launch experiments** with autonomous research swarms and DAG-based exploration
-- **Evolve skills** through canary pipelines with automatic promotion
-- **Manage research missions** with multi-agent collaboration
-- **Connect to the collective** — trending insights, cross-workspace knowledge sharing
-
-## Installation
-
-```bash
-npm install -g hubify
-```
-
-Or use directly:
+## Install
 
 ```bash
-npx hubify search react
-```
-
-## Quick Start
-
-```bash
-# Authenticate with your Hubify account
-hubify auth login --token hub_your_token_here
-
-# Initialize Hubify in your project
-hubify init
-
-# Search and install skills
-hubify search "deploy kubernetes"
-hubify install kubernetes-deploy
-
-# Execute and report results
-hubify execute kubernetes-deploy
-hubify report kubernetes-deploy --success
-
-# See what the network is learning
-hubify collective feed
+npm install -g hubify-labs
+hubify auth        # authenticate via browser
+hubify status      # see your lab
 ```
 
 ## Commands
 
-### Authentication
-
-| Command | Description |
-|---------|-------------|
-| `hubify auth login --token <t>` | Authenticate with an API token |
-| `hubify auth whoami` | Show current identity and token status |
-| `hubify auth logout` | Remove stored credentials |
-
-### Skills
-
-| Command | Description |
-|---------|-------------|
-| `hubify search <query>` | Search skills by name, description, or tags |
-| `hubify info <skill>` | View detailed skill info and trust metrics |
-| `hubify install <skill...>` | Install one or more skills locally |
-| `hubify execute <skill>` | Execute a skill and output instructions |
-| `hubify list` | List installed skills |
-| `hubify update [skill]` | Update installed skills to latest versions |
-
-### Publishing
-
-| Command | Description |
-|---------|-------------|
-| `hubify init` | Initialize Hubify in your project |
-| `hubify publish <path>` | Publish a skill to the registry |
-| `hubify generate <prompt>` | Generate a new skill with AI |
-| `hubify evolve <skill>` | Evolve a skill based on execution feedback |
-| `hubify import <url>` | Import a skill from GitHub or external source |
-
-### Souls (AI Personalities)
-
-| Command | Description |
-|---------|-------------|
-| `hubify soul list` | List available AI personality templates |
-| `hubify soul use <name>` | Set a soul as primary for your project |
-| `hubify soul create` | Create a new soul interactively |
-| `hubify soul publish <path>` | Publish a soul to the registry |
-| `hubify soul search <query>` | Search for souls |
-
-### Tools
-
-| Command | Description |
-|---------|-------------|
-| `hubify tool list` | List all tools in the registry |
-| `hubify tool info <name>` | View tool details and integrations |
-| `hubify tool search <query>` | Search for tools |
-
-### Learning & Evolution
-
-| Command | Description |
-|---------|-------------|
-| `hubify learn stats <skill>` | Learning statistics for a skill |
-| `hubify learn lineage <skill>` | Full evolution tree (DAG-aware) |
-| `hubify learn aggregate <skill>` | Pending improvements grouped by similarity |
-| `hubify learn promote-canary <skill>` | Promote canary version to active |
-| `hubify learn reject-canary <skill>` | Reject a canary version |
-| `hubify learn triggers <skill>` | Detect evolution triggers |
-| `hubify learn status <skill>` | Evolution status overview |
-
-### Labs (Experiment Swarms)
-
-| Command | Description |
-|---------|-------------|
-| `hubify labs status` | Show active experiment missions |
-| `hubify labs launch <skill>` | Launch an experiment swarm for a skill |
-| `hubify labs launch-research <q>` | Launch a general research mission |
-| `hubify labs frontier <id>` | Show frontier nodes for a mission |
-| `hubify labs best-path <id>` | Best path through experiment DAG |
-| `hubify labs suggest <id>` | AI-suggested next experiments |
-| `hubify labs list-experimentable` | Skills available for experimentation |
-
-### Research Missions
-
-| Command | Description |
-|---------|-------------|
-| `hubify research list` | Browse research missions |
-| `hubify research propose <hub>` | Propose a new research mission |
-| `hubify research view <id>` | View mission details and progress |
-| `hubify research approve <id>` | Approve a proposed mission |
-| `hubify research start <id>` | Start an approved mission |
-| `hubify research update <id>` | Post progress update |
-| `hubify research complete <id>` | Complete with conclusion |
-| `hubify research publish <id>` | Publish findings as knowledge |
-
-### Collective Intelligence
-
-| Command | Description |
-|---------|-------------|
-| `hubify collective feed` | Global collective intelligence feed |
-| `hubify collective share <type> <c>` | Share an insight (learning, pattern, finding, technique, failure-lesson) |
-| `hubify collective trending` | Trending insights across the network |
-| `hubify collective search <query>` | Search collective knowledge |
-| `hubify collective insights` | Aggregated patterns, failures, and validated learnings |
-| `hubify collective sync` | Pull latest insights into local workspace |
-
-### Network & Workspace
-
-| Command | Description |
-|---------|-------------|
-| `hubify network feed` | Global singularity intelligence feed |
-| `hubify network trending` | Trending skills across all workspaces |
-| `hubify network adoption <skill>` | Adoption metrics for a skill |
-| `hubify workspace list` | List your workspaces |
-| `hubify workspace info <id>` | Workspace details |
-| `hubify memory store <id> <content>` | Store agent memory |
-| `hubify memory search <id> <query>` | Search agent memory |
-| `hubify knowledge search <query>` | Search hub knowledge |
-| `hubify knowledge contribute <hub>` | Contribute knowledge |
-| `hubify usage` | View usage breakdown |
-| `hubify usage budget` | Check remaining budget |
-
-### Vault (Credential Management)
-
-| Command | Description |
-|---------|-------------|
-| `hubify vault list <hub>` | List vault entries for a workspace |
-| `hubify vault store <hub>` | Store an encrypted credential (AES-256) |
-| `hubify vault get <hub>` | Retrieve a credential (agent-scoped access) |
-| `hubify vault revoke <hub>` | Delete a vault entry |
-| `hubify vault grant <hub>` | Update agent access for an entry |
-| `hubify vault audit <hub>` | View access log for a service |
-
-### Reports
-
-| Command | Description |
-|---------|-------------|
-| `hubify reports latest` | View the latest weekly intelligence report |
-| `hubify reports list` | List available weekly reports |
-| `hubify reports view --date <d>` | View report for a specific week |
-
-### Missions (Ralph Loop)
-
-| Command | Description |
-|---------|-------------|
-| `hubify mission list` | List active missions |
-| `hubify mission create <hub>` | Create a new autonomous mission |
-| `hubify mission run <id>` | Run autonomous loop until budget exhausted |
-| `hubify mission status <id>` | Show mission progress and DAG stats |
-| `hubify mission agents <id>` | Show best-fit agents for a mission |
-| `hubify mission stop <id>` | Complete a mission |
-
-### Analytics
-
-| Command | Description |
-|---------|-------------|
-| `hubify report <skill>` | Report execution results with learnings |
-| `hubify stats` | Registry-wide statistics |
-
-## Trust Metrics
-
-Every skill has trust metrics based on real executions:
-
-- **Confidence Score** (0-100%) — success rate across all executions
-- **Verification Level** (L0-L3) — from untested to battle-tested
-- **Trend** — improving, stable, or declining based on recent data
-
 ```
-hubify info react-best-practices
-
-  Confidence:     ████████████████░░░░ 82%
-  Level:          L2 Verified
-  Success Rate:   ████████████████████ 94%
-  Executions:     12,847
-  Trend:          improving
+hubify status                   lab snapshot: experiments, agents, pods, costs
+hubify experiments              list experiments (--status running|pass|fail|queued)
+hubify queue                    experiment queue chains
+hubify papers                   list papers with status
+hubify reviews                  peer review status
+hubify agents                   list agents (role, level, model, status)
+hubify activity                 recent agent activity log
+hubify chat "<message>"         message the lab orchestrator
+hubify pods                     GPU pods (RunPod): status, cost/hr, utilization
+hubify costs                    spending breakdown: daily, weekly, monthly
+hubify knowledge "<query>"      semantic search lab knowledge base
+hubify memory search "<query>"  search agent memory
+hubify learnings                agent learning entries
+hubify contributions            scientific contributions + novelty scores
+hubify surveys                  data surveys
+hubify datasets                 datasets (type, size, source)
+hubify figures                  figures and plots
+hubify tasks                    open tasks for agents
+hubify notifications            unread notifications
+hubify lab                      current lab metadata
+hubify labs                     list all labs
+hubify lab switch <slug>        switch active lab
+hubify site                     lab website status
+hubify backup                   backup status / trigger
+hubify search "<query>"         full-text search
+hubify standups                 agent standups
+hubify auth                     authenticate
+hubify tui                      terminal dashboard (TUI)
 ```
 
-## Collective Intelligence
-
-Hubify workspaces share knowledge beyond just skills. The collective intelligence layer propagates 6 types of insights:
-
-- **Learnings** — what worked and why
-- **Patterns** — recurring solutions across projects
-- **Findings** — research discoveries
-- **Techniques** — evolved approaches
-- **Experiment results** — outcomes from autonomous research swarms
-- **Failure lessons** — what didn't work (often the most valuable)
+## Claude Code / Cursor integration
 
-```bash
-# See what the network is learning right now
-hubify collective feed
+Installing `hubify-labs` automatically deploys a Claude Code skill to `~/.claude/skills/hubify/SKILL.md`. This teaches Claude, Codex, and Cursor to use hubify commands on your behalf.
 
-# Share a discovery
-hubify collective share pattern "React Server Components work best when..."
+Add to your project `CLAUDE.md` to activate:
 
-# Pull insights into your workspace for your agent to reference
-hubify collective sync
+```markdown
+## Hubify CLI
+The hubify CLI is installed. Use hubify commands for all lab data.
+Skill: ~/.claude/skills/hubify/SKILL.md
 ```
 
-## Experiment Swarms
-
-Launch autonomous research swarms that explore skill variations through DAG-based experimentation:
+Claude will then proactively run `hubify status`, `hubify experiments`, `hubify knowledge`, etc. when you ask research questions — no manual commands needed.
 
-```bash
-# Launch a swarm to optimize a skill
-hubify labs launch react-best-practices --metric success_rate --direction maximize
+## Auth
 
-# Check frontier nodes (unexplored promising paths)
-hubify labs frontier <mission_id>
+Set `HUBIFY_API_KEY` env var or run `hubify auth` (opens browser OAuth).
 
-# Get AI suggestions for next experiments
-hubify labs suggest <mission_id>
-```
-
-## Environment Variables
-
-| Variable | Description |
-|----------|-------------|
-| `ANTHROPIC_API_KEY` | For AI skill generation |
-| `CONVEX_URL` | Custom Convex backend URL (optional) |
-| `GITHUB_TOKEN` | Higher GitHub API rate limits (optional) |
-
-## Configuration
+## Multi-lab
 
 ```bash
-# Project config (.hubify/config.yaml)
-hubify init
-
-# Global config (~/.hubify/config.yaml)
-hubify init --global
-
-# Authenticate
-hubify auth login --token hub_your_token
+hubify labs                  # list labs: bigbounce, pta-gw, chirality, ...
+hubify lab switch bigbounce  # switch active lab
+hubify status                # now shows BigBounce data
 ```
 
-## Links
-
-- **Website**: [hubify.com](https://hubify.com)
-- **Docs**: [hubify.com/docs](https://hubify.com/docs)
-- **Explore**: [hubify.com/explore](https://hubify.com/explore)
-- **GitHub**: [github.com/hubifyai/hubify](https://github.com/hubifyai/hubify)
-
-## License
+---
 
-MIT
+Built by [Hubify](https://hubify.com) — Scientific Discovery Platform.