Vive — AI skill for Claude Code
Smart MCP harness for AI agents — auto-memory, context compression, and task orchestration in a single daemon.
How to install Vive
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open wess/vive and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Vive does
Smart MCP harness for AI agents — auto-memory, context compression, and task orchestration in a single daemon.
Alternatives in AI
- Open Multi Agent — TypeScript multi-agent orchestration engine — one runTeam() call from goal to result 5.8k ★
- Paritok 4b V1 — Non-destructive compression gateway for AI coding agents 1.4k ★
- Agenvoy — Self-hosted AI agent harness in a single Go binary — writes, sandbox-tests and repairs its own tools, and lets 474 ★
README
Vive
A smart harness server for AI agent orchestration. Manages context, memory, and task routing so agents stay focused and token-efficient.
What It Does
Vive sits between AI clients and model providers as invisible infrastructure. Once set up, it automatically:
- Remembers decisions — detects and persists project decisions across sessions
- Tracks file changes — knows what was created and modified
- Injects context — at session start, surfaces relevant memories so you never re-explain
- Manages tasks — create, run, cancel, and monitor tasks with execution strategies
- Orchestrates agents — register agents, assign tasks, handoff context between them
- Shows you what's happening — live ncurses TUI dashboard with 5 panels
Quick Start
Install (macOS and Linux)
curl -fsSL https://raw.githubusercontent.com/wess/vive/main/install.sh | sh
vive init
Install and configure Claude Code in one shot:
curl -fsSL https://raw.githubusercontent.com/wess/vive/main/install.sh | sh -s -- --setup
Options: `--version vX.Y.Z`, `--prefix /path`, `--no-deps`. Run with `--help` for details.
Homebrew
brew install wess/tap/vive
vive init
Build From Source
Requires a C compiler, [Goose](https://github.com/wess/goose), and `sqlite3` + `cjson` + `ncurses`:
goose build # ./build/debug/vive
goose build --release
goose install
vive init --global # enable for all projects
How It Works
Seamless Integration
`vive init` configures everything in one shot:
- Registers Vive as an MCP server (Claude Code can call all 17 tools directly)
- Installs hooks that automatically capture events:
- UserPromptSubmit — stores prompts, detects decisions
- PostToolUse — tracks file changes, build commands
- SessionStart — injects relevant context as a system reminder + updates CLAUDE.md
- Stop — creates session summary, prunes expired memories
- Creates a
/viveslash
Related Skills
TOMC
TOMC Memory: open-source context compression and task memory for Claude Code, Codex and LLM APIs. CPU-only pre
CodeForge
CodeForge — a terminal AI coding assistant in Python (Claude Code-style). Pluggable ReAct agent loop, extensib
Openproxy
Smart AI proxy router that connects Claude Code, Cursor, Codex, Copilot, and 50+ coding tools to 40+ AI provid
Distil
Measure what context compression actually costs an LLM agent: where a session's tokens go, and whether a rewri
Cost Optimizer
git clone https://github.com/fullstackcrew-alpha/skill-cost-optimizer Save 60-80% on AI token costs with smart
Llmslim
Long-term memory for AI agents with no LLM call: an MCP server and Claude Code plugin, plus local-first contex
Related Agents
Claude Harness Engineer
Changes Claude Code harness configuration — settings.json, statusline.sh, CLAUDE.md global rules, hooks, rules
Forge Master
Forge Task Creation Master - Creates optimized single-group tasks in Forge MCP with comprehensive @ context lo
Context Memory Manager
Context Window & Token Optimization Specialist. Manages token budgets, compresses prompts (RTK/Caveman), imple