Rank #50 · on radar since 2026-07-07
metaharness
🛠️ The meta-harness for AI agents — scaffold your own focused, branded agent harness with its own npx CLI, MCP server, memory, learning loop, and witness-signed releases. Works with Claude Code, Codex, pi.dev, Hermes, OpenClaw, and RVM (hardware-isolated sandbox).
metaharness is a AI agent on the Cresting radar since 2026-07-07, currently ranked #50 of 2371 tracked tools with a momentum score of 65.2. It has 63,472 GitHub stars (+104 in the last 7 days). Momentum measures how fast a tool is rising, not how big it is; the score recomputes hourly from public signals.
Momentum
65.2
24h· new7d· new
Why it's ranked
Every score decomposes into published factors — the same math for every tool, paid or not. Read the methodology →
| Velocity (weighted, cohort-normalized) | 0.982 |
| Recency boost | 0.174 |
| Signal decay | 0.991 |
| Corroboration | 1.000 |
| Quality gate | 1.000 |
Raw signals (30 days)
github · forks+19 in window · 7,480 latest · 2 snapshots
github · stars+117 in window · 63,485 latest · 2 snapshots
Related ai agents
- motion-anything✨ The agentic motion layer — an open-source, chat-native motion engine. Describe the feeling; your AI ships the animation.#1 · 230★
- Vibe-ResearchVibe-Research: Your Personal Trading Research Agent · A股/美股/港股 的个人投研 Agent：每日复盘、资讯雷达、个股数据、板块中心、我的持仓、研究记录。Vibe-Research 把数据和功能配齐，由你自己的 AI 驱动投资研究。#2 · 479★
- openscienceThe open-source AI workbench for scientific research#3 · 1,466★
- engramEvidence-based learning engine for Claude Code — first-principles curricula, free-recall verification with receipts, FSRS-scheduled memory, and explorable artifacts. Learn anything; keep it.#4 · 386★
- T3MP3STautonomous red teaming platform; multi-agent offensive-security meta-harness#5 · 3,426★