# INK-Q/agy-fast > A local MCP plugin for Google Antigravity (CLI/Desktop/IDE) that speeds up the coding agent by batching repeated tool calls and cutting LLM round-trips and context size. It does not touch the native UI, diffs, or built-in tools. - Magnitude: 5.6 out of 10 — Early signal - Stars: 192 total · +29 stars measured 2026-11-21, 00:34–03:57 UTC, ≈ 157 by evening - Star trust: star growth looks organic - Category: AI agents · Language: TypeScript · License: MIT · Created: 2026-11-19 · Last push: 2026-11-20 - GitHub: https://github.com/INK-Q/agy-fast · Page: https://gitnova.dev/en/r/INK-Q/agy-fast ## Useful for - Install the plugin with one command and enable it via agy plugin enable - Cut token usage and LLM round-trips during long coding-agent sessions - Run the local benchmark on your own transcripts via node dist/benchmark.mjs analyze ## Why it’s here - Star-counter measurements on 2026-11-21 (UTC), 00:34–03:57: 163 → 192 stars (+29). This is the change over that interval. - Estimated end-of-day forecast: about +157 stars, using observed gains and the previous day. - The repository is 2 days old and already has 192 stars. With less than two weeks of history, there's no usual pace to compare the spike against yet. - Top new repositories this week: #93. - About 9 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: 16 - Issues and pull requests: 3 - Watchers: 1 - Average over the last week: 114 per day - Usual pace: too little history (under two weeks) - Stars in the last hour (measured): 7 - Latest release: v0.2.3 (2026-11-20) ## Stars per day, last 3 days (oldest → newest, today is partial) 2026-11-19 … 2026-11-21: 34, 151, 29 ## Spotted in now - Top new repositories this week: #93 ## More in this category 1. **simplexai-labs/linger** — 7.7 · Breakout · AI agents · TypeScript · +104 stars measured 2026-11-21, 00:34–03:57 UTC, ≈ 726 by evening An open-source personal AI agent that runs fully locally on your computer: it understands context across files and apps, keeps memory in local wiki pages, and proactively completes tasks. Data and memories stay on your device. Full card: https://gitnova.dev/en/r/simplexai-labs/linger.md 2. **bookwjitchieni1986/blender-ai-mcp-addon** — 7.0 · Breakout · AI agents · +42 stars measured 2026-11-21, 00:34–03:57 UTC, ≈ 679 by evening AI assistant for Blender with a visual chat workspace and built-in MCP server: inspects scenes, creates objects, edits materials and runs multi-step tasks via connected LLMs. Full card: https://gitnova.dev/en/r/bookwjitchieni1986/blender-ai-mcp-addon.md 3. **nataphong-lab/Open-Source-Face-Recognition-Ultimate** — 6.8 · Early signal · AI agents · Python · +29 stars measured 2026-11-21, 00:34–03:57 UTC, ≈ 221 by evening Open-source Python SDK for face recognition: detection, landmarks, embedding extraction and face comparison. All processing runs locally on Windows and Linux without sending biometric data to external servers. Full card: https://gitnova.dev/en/r/nataphong-lab/Open-Source-Face-Recognition-Ultimate.md 4. **Tracer-Cloud/opensre** — 6.7 · Breakout · AI agents · Python · +67 stars measured 2026-11-21, 00:34–03:57 UTC, ≈ 401 by evening Open-source framework for building AI SRE agents that investigate production incidents across logs, metrics, traces and deploys on your own infrastructure. Full card: https://gitnova.dev/en/r/Tracer-Cloud/opensre.md 5. **mdiri-ai/SoduWork** — 6.2 · Breakout · AI agents · JavaScript · +90 stars measured 2026-11-21, 00:34–03:57 UTC, ≈ 1,046 by evening Self-hosted platform for team workflows with AI agents: requests carry acceptance criteria, a separate model reviews the work, and a human signs off. Supports multiple harnesses (Pi, OpenCode, Codex, etc.). Full card: https://gitnova.dev/en/r/mdiri-ai/SoduWork.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 2026-11-21 04:07 UTC, updated every 30 minutes.