# mcpdelta/mcpdelta > A desktop app that proxies between AI apps and their MCP servers, letting the model run a whole task as one program instead of one tool call per step, with an activity log and one-click undo. It aims to cut tokens, cost and time on multi-step MCP tasks; the app itself is not yet released and this repo only presents it. - Magnitude: 5.6 out of 10 — Early signal - Stars: 79 total · +0 stars measured 2026-10-09, 00:05–01:30 UTC, ≈ 68 by evening - Star trust: star growth looks organic - Category: AI agents · Created: 2026-10-07 · Last push: 2026-10-08 - GitHub: https://github.com/mcpdelta/mcpdelta · Homepage: https://mcpdelta.com · Page: https://gitnova.dev/en/r/mcpdelta/mcpdelta ## Useful for - Cut token usage on multi-step MCP tasks in Claude Code or Cursor by switching apps to Delta MCP - Roll back an agent's file or data changes with the one-click undo in the Activity log - Inspect every tool call an AI app makes through its existing MCP servers before and after each change ## Why it’s here - Star-counter measurements on 2026-10-09 (UTC), 00:05–01:30: 79 → 79 stars (+0). This is the change over that interval. - Estimated end-of-day forecast: about +68 stars, using observed gains and the previous day. - The repository is 2 days old and already has 79 stars. With less than two weeks of history, there's no usual pace to compare the spike against yet. - Top new repositories this week: #185. - About 200 forks a day — people are taking the code. ## Star trust Star growth looks organic. Star-trust labels are heuristics based on the repository’s behavior, not a check of every stargazer. ## Numbers - Forks: 11 - Issues and pull requests: 0 - Watchers: 0 - Average over the last week: 49 per day - Usual pace: too little history (under two weeks) - Stars in the last hour (measured): 0 ## Stars per day, last 6 days (oldest → newest, today is partial) 2026-10-04 … 2026-10-09: 0, 0, 0, 0, 79, 0 ## Spotted in now - Top new repositories this week: #185 ## More in this category 1. **morluto/rea** — 9.3 · Breakout · AI agents · TypeScript · +1,051 stars measured 2026-10-09, 00:04–01:30 UTC, ≈ 10,366 by evening A reverse-engineering tool that lets AI agents inspect apps and native binaries: it connects agents to Hopper or Ghidra via MCP and CLI, helping explain how a feature works and rebuild it in your own project. Full card: https://gitnova.dev/en/r/morluto/rea.md 2. **docker/docker-agent** — 7.5 · Breakout · AI agents · Go · +6 stars measured 2026-10-09, 00:04–01:30 UTC, ≈ 307 by evening Docker CLI plugin for building and running AI agents from a declarative YAML config with tools, MCP servers, and multi-agent orchestration. Lets you assemble agent teams without writing code and share them via OCI registry. Full card: https://gitnova.dev/en/r/docker/docker-agent.md 3. **zhongerxin/iPhone-use** — 6.9 · Early signal · AI agents · Python · +44 stars measured 2026-10-09, 00:05–01:30 UTC, ≈ 408 by evening Lets Codex control a real iPhone over USB via WebDriverAgent: it installs and signs WDA, exposes MCP tools for app navigation, taps, typing and list collection, and shows a live screen widget. Full card: https://gitnova.dev/en/r/zhongerxin/iPhone-use.md 4. **thedotmack/claude-mem** — 6.8 · Breakout · AI agents · TypeScript · +45 stars measured 2026-10-09, 00:05–01:30 UTC, ≈ 604 by evening Persistent memory system for Claude Code and other agents: it captures agent actions via lifecycle hooks, compresses them into semantic summaries, and injects relevant context into future sessions. Full card: https://gitnova.dev/en/r/thedotmack/claude-mem.md 5. **thesysdev/open-intelligent-ui** — 5.9 · Early signal · AI agents · TypeScript · +8 stars measured 2026-10-09, 00:07–01:30 UTC, ≈ 134 by evening An open-source implementation of ChatGPT's "Intelligent UI" built on OpenUI: the model streams OpenUI Lang and the app renders an interactive answer from custom travel components (map, gallery, itinerary). Full card: https://gitnova.dev/en/r/thesysdev/open-intelligent-ui.md --- Magnitude (0–10) measures how fast and how unusually interest in a repository is growing right now. It is not a quality score. Days are UTC. “So far today” is a fact; “expected by the end of the day” is a forecast. Summaries and use cases are written by an LLM (DeepSeek V4.1 Flash) from the README and may be inaccurate: verify specific claims (benchmarks, speed, hardware) in the repository itself. Data as of 2026-10-09 01:41 UTC, updated every 30 minutes.