Skip to content
Back to search

agy-fast

Non-invasive MCP accelerator plugin for Google Antigravity (CLI/Desktop/IDE): cuts LLM turns (-34%), token quota (-31%), and upload traffic (-40%) without affecting native UI or diffs.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/ink-q--agy-fast/badge.svg)](/listings/ink-q--agy-fast)

HTML

<a href="/listings/ink-q--agy-fast"><img src="/api/listings/ink-q--agy-fast/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

These steps run a script from the network that findsafeskills does not scan.

**Requirements:** **Node.js >= 20** (no `npm install` needed — pre-built, unminified zero-dependency ESM bundles are included in `dist/`). Optional: [`ripgrep` (`rg`)](https://github.com/BurntSushi/ripgrep) for faster code search (automatically falls back to pure Node.js when absent).

One-Command Install (Recommended)

Installs runtime files to `~/.gemini/config/plugins/agy-fast`, runs `agy plugin validate --strict` and `agy plugin enable agy-fast` when `agy` is installed, and verifies both the MCP server and `SessionStart` hook live:

**macOS / Linux:**

curl -fsSL https://raw.githubusercontent.com/INK-Q/agy-fast/main/install.sh | sh

**Windows (PowerShell):**

irm https://raw.githubusercontent.com/INK-Q/agy-fast/main/install.ps1 | iex

> [!TIP] > **Activating in Desktop 2.0 / IDE:** Open **Customizations → Installed** in Antigravity, confirm `agy-fast` is enabled, and start a **new session** to load the `SessionStart` hook and MCP tools.

<details> <summary>More installation options (local clone, direct git clone, <code>agy</code> CLI, per-repo team install, or ask Antigravity)</summary>

Local Clone (Any OS)

git clone https://github.com/INK-Q/agy-fast.git
cd agy-fast
node scripts/install.mjs install

Direct Clone into Antigravity Plugins Directory (No Script Needed)

Antigravity automatically discovers plugins placed in `~/.gemini/config/plugins/`:

# macOS / Linux
git clone https://github.com/INK-Q/agy-fast.git ~/.gemini/config/plugins/agy-fast
# Windows (PowerShell)
git clone https://github.com/INK-Q/agy-fast.git "$env:USERPROFILE\.gemini\config\plugins\agy-fast"

Native Antigravity CLI (`agy`)

agy plugin install /path/to/agy-fast
agy plugin enable agy-fast

Per-Repository Installation (Team Sharing)

Install only for a single repository (writes to `<repo>/.agents/plugins/agy-fast`):

node scripts/install.mjs install --project /path/to/your-repo

Ask Antigravity Directly

Paste this prompt into Antigravity:

Clone https://github.com/INK-Q/agy-fast.git to ~/.gemini/config/plugins/agy-fast, run `node ~/.gemini/config/plugins/agy-fast/scripts/install.mjs doctor`, and enable the agy-fast plugin.

</details>

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 318 lines · 2 findings marked
  1. <div align="center">
  2. <picture>
  3. <source media="(prefers-color-scheme: dark)" srcset="assets/logo-light.svg" />
  4. <source media="(prefers-color-scheme: light)" srcset="assets/logo-dark.svg" />
  5. <img src="assets/logo.svg" alt="agy-fast logo" width="80" />
  6. </picture>
  7. # agy-fast
  8. **Latency, Token-Quota & Tool-Call Accelerator Plugin for Google Antigravity**
  9. [![Version](https://img.shields.io/github/v/tag/INK-Q/agy-fast?label=version&sort=semver&style=flat-square&color=38BDF8&labelColor=0F172A)](CHANGELOG.md)
  10. [![Node.js >= 20](https://img.shields.io/badge/node-%3E%3D20-22C55E?style=flat-square&logo=node.js&logoColor=white&labelColor=0F172A)](#quick-start)
  11. [![Runtime Dependencies: 0](https://img.shields.io/badge/runtime%20deps-0-38BDF8?style=flat-square&labelColor=0F172A)](#how-it-works)
  12. [![Platforms](https://img.shields.io/badge/platform-macOS%20%7C%20Linux%20%7C%20Windows-64748B?style=flat-square&labelColor=0F172A)](#quick-start)
  13. [![MIT License](https://img.shields.io/badge/license-MIT-22C55E?style=flat-square&labelColor=0F172A)](LICENSE)
  14. English · [简体中文](README.zh-CN.md) · [Quick Start](#quick-start) · [Benchmarks](#real-world-benchmarks) · [MCP Tools](#the-4-mcp-tools) · [Analysis Report](docs/ANALYSIS_REPORT.md) · [Changelog](CHANGELOG.md)
  15. </div>
  16. <br />
  17. **`agy-fast`** is a zero-dependency, policy-compliant local [MCP](https://modelcontextprotocol.io/) plugin for **Google Antigravity** (CLI `agy`, Desktop 2.0, and IDE). It cuts agent wall-clock latency, billable token quota, and upload traffic by consolidating high-frequency searches, multi-language symbol outlines, atomic multi-file edits, and build/test checks into batched local calls—and injecting a sub-50 ms workspace snapshot once per session—without altering Antigravity's native UI, diff viewer, modified-files list, checkpoints, or safety checks.
  18. | LLM Turns / 10 Actions | Est. Billable Quota | Tool-Return Upload | Spills & Native UI Impact |
  19. | :---: | :---: | :---: | :---: |
  20. | **-33.7%**<br><sub>`-38.2%` ≥150k ctx · `-58.8%` subagents</sub> | **-31.5%**<br><sub>`-36.5%` ≥150k ctx · `-54.0%` subagents</sub> | **-39.7%**<br><sub>`-53.7%` uncached fresh input</sub> | **0**<br><sub>`≤8.8 KB` inline cap · Native diff & UI untouched</sub> |
  21. ## Quick Start
  22. **Requirements:** **Node.js >= 20** (no `npm install` needed — pre-built, unminified zero-dependency ESM bundles are included in `dist/`). Optional: [`ripgrep` (`rg`)](https://github.com/BurntSushi/ripgrep) for faster code search (automatically falls back to pure Node.js when absent).
  23. ### One-Command Install (Recommended)
  24. Installs runtime files to `~/.gemini/config/plugins/agy-fast`, runs `agy plugin validate --strict` and `agy plugin enable agy-fast` when `agy` is installed, and verifies both the MCP server and `SessionStart` hook live:
  25. **macOS / Linux:**
  26. ```bash
  27. curl -fsSL https://raw.githubusercontent.com/INK-Q/agy-fast/main/install.sh | sh
    • Note: Pipes a downloaded script directly into a shell (curl | sh / wget | bash)
    • Line 41 on GitHub
  28. ```
  29. **Windows (PowerShell):**
  30. ```powershell
  31. irm https://raw.githubusercontent.com/INK-Q/agy-fast/main/install.ps1 | iex
  32. ```
  33. > [!TIP]
  34. > **Activating in Desktop 2.0 / IDE:** Open **Customizations → Installed** in Antigravity, confirm `agy-fast` is enabled, and start a **new session** to load the `SessionStart` hook and MCP tools.
  35. <details>
  36. <summary>More installation options (local clone, direct git clone, <code>agy</code> CLI, per-repo team install, or ask Antigravity)</summary>
  37. #### Local Clone (Any OS)
  38. ```bash
  39. git clone https://github.com/INK-Q/agy-fast.git
  40. cd agy-fast
  41. node scripts/install.mjs install
  42. ```
  43. #### Direct Clone into Antigravity Plugins Directory (No Script Needed)
  44. Antigravity automatically discovers plugins placed in `~/.gemini/config/plugins/`:
  45. ```bash
  46. # macOS / Linux
  47. git clone https://github.com/INK-Q/agy-fast.git ~/.gemini/config/plugins/agy-fast
  48. ```
  49. ```powershell
  50. # Windows (PowerShell)
  51. git clone https://github.com/INK-Q/agy-fast.git "$env:USERPROFILE\.gemini\config\plugins\agy-fast"
  52. ```
  53. #### Native Antigravity CLI (`agy`)
  54. ```bash
  55. agy plugin install /path/to/agy-fast
  56. agy plugin enable agy-fast
  57. ```
  58. #### Per-Repository Installation (Team Sharing)
  59. Install only for a single repository (writes to `<repo>/.agents/plugins/agy-fast`):
  60. ```bash
  61. node scripts/install.mjs install --project /path/to/your-repo
  62. ```
  63. #### Ask Antigravity Directly
  64. Paste this prompt into Antigravity:
  65. ```text
  66. Clone https://github.com/INK-Q/agy-fast.git to ~/.gemini/config/plugins/agy-fast, run `node ~/.gemini/config/plugins/agy-fast/scripts/install.mjs doctor`, and enable the agy-fast plugin.
  67. ```
  68. </details>
  69. ## How It Works
  70. Empirical analysis of **55 real Antigravity sessions** (**8,446 LLM turns**, **17,302 tool calls**, **55.25 hours** of single-agent active time; see [docs/ANALYSIS_REPORT.md](docs/ANALYSIS_REPORT.md)) revealed that slow tasks are rarely limited by token generation speed. Instead, four structural bottlenecks dominate wall-clock time:
  71. | Measured Bottleneck | Empirical Cost | How `agy-fast` Eliminates It |
  72. | :--- | :--- | :--- |
  73. | **Per-turn TTFT floor** | `~6.1 s` median TTFT (`79%` of a median turn; `68.1%` of turns emit `<700` tokens) | Batches an average of **4.89** logical actions per MCP call (`5.95×` in subagents), cutting unnecessary round-trip turns (`~6–8 s` each) |
  74. | **Serial checks & fragmented ops** | `52.5%` of `edit_file` calls are same-file splits; `46.3%` of `run_command` calls are read-only searches (`~2.0 s` median vs `2.1 ms` local `rg`) | Consolidates up to 20 searches, 50 file edits + `verify`, or 12 build/test commands inside one persistent local Stdio MCP server |
  75. | **Cold-start exploration** | First 3 turns average **49.2 seconds** inspecting directory layout, git state, and scripts | `SessionStart` hook injects a `≤1.5 KB` snapshot (git state, 2-level tree, multi-module scripts) once in `~35–50 ms` |
  76. | **Context bloat & cache decay** | `read_file` accounts for **81.9%** of tool-returned bytes; `>60 s` gaps spike cache misses from `1.4%` to `15.9%–93.8%` | Symbol outlines first + parallel targeted `read_file`, hard `≤8.8 KB` UTF-8 inline cap (zero `output.txt` disk spills), and multimodal isolation rules |
  77. ### Hybrid Execution Architecture
  78. `agy-fast` never replaces fast native reads. It combines batched local MCP tools with Antigravity's locally allowlisted `read_file` (`~7 ms`, `46 KB/call` inline, zero remote check):
  79. <p align="center">
  80. <picture>
  81. <source media="(prefers-color-scheme: dark)" srcset="assets/architecture-dark.svg" />
  82. <source media="(prefers-color-scheme: light)" srcset="assets/architecture-light.svg" />
  83. <img alt="agy-fast Hybrid Execution Architecture" src="assets/architecture-light.svg" width="100%" />
  84. </picture>
  85. </p>
  86. ### Plugin Components
  87. | Component | Path | Role |
  88. | :--- | :--- | :--- |
  89. | **Plugin Manifest** | [`plugin.json`](plugin.json) | Standard Antigravity manifest; passes `agy plugin validate --strict` |
  90. | **Local Stdio MCP Server** | [`mcp_config.json`](mcp_config.json) → [`dist/mcp-server.mjs`](dist/mcp-server.mjs) | Persistent local process exposing 4 composite tools under `agy-fast_fast` |
  91. | **Startup Snapshot Hook** | [`hooks.json`](hooks.json) → [`dist/session-start.mjs`](dist/session-start.mjs) | Runs once on `SessionStart` (`~35–50 ms`), emitting a `≤1.5 KB` workspace snapshot |
  92. | **Routing & Hygiene Rules** | [`rules/AGENTS.md`](rules/AGENTS.md) | Compact (`~2.2 KB`) routing guide with inlined schemas (zero schema-read turns) |
  93. | **Background Auto-Updater** | [`dist/updater.mjs`](dist/updater.mjs) | Detached 12 h release checker with staged 4-tool MCP smoke verification |
  94. | **Offline Benchmark CLI** | [`dist/benchmark.mjs`](dist/benchmark.mjs) | Local `transcript.jsonl` analyzer and before/after A/B comparator |
  95. ## Real-World Benchmarks
  96. Across **10 days of sanitized real coding transcripts** (**14 baseline main sessions / 5,771 turns** vs. **7 `agy-fast` main sessions / 901 turns**, plus **104 baseline vs. 4 `agy-fast` subagent sessions**; all repository names, paths, and identifiers stripped), each `agy-fast` call batched an average of **4.89 logical operations** (`5.95×` in subagents):
  97. | Metric *(normalized per logical action)* | Main Sessions (Baseline) | Main Sessions (`agy-fast`) | Delta | `≥150k` Long Context | Subagents |
  98. | :--- | ---: | ---: | ---: | ---: | ---: |
  99. | **LLM turns / 10 actions** | 5.23 | **3.47** | **-33.7%** | **-38.2%** | **-58.8%** |
  100. | **Upstream requests / 10 actions** *(LLM + remote checks)* | 9.82 | **6.05** | **-38.4%** | **-42.6%** | **-62.0%** |
  101. | **Prompt Cache hit rate** *(miss spikes)* | 89.6% *(7.5%)* | **94.0%** *(4.8%)* | **+4.4 pp** | **95.0%** *(3.3%)* | 92.4% *(2.8%)* |
  102. | **Uncached fresh input / action** | 10,413 tok | **4,822 tok** | **-53.7%** | **-61.5%** | **-51.3%** |
  103. | **Est. billable quota / action** *(1 : 0.1 : 5)* | 22,458 eq | **15,387 eq** | **-31.5%** | **-36.5%** | **-54.0%** |
  104. | **Tool-return upload / action** *(actual local bytes)* | 5.06 KB | **3.05 KB** | **-39.7%** | **-42.6%** | **-13.3%** |
  105. | **Incremental text upload / action** *(fresh input est.)* | 35.6 KB | **16.5 KB** | **-53.7%** | **-61.6%** | **-51.3%** |
  106. | **Full-context text upload / action** *(stateless resend est.)* | 342.2 KB | **276.2 KB** | **-19.3%** | **-21.9%** | **-58.9%** |
  107. **Production hardening highlights (`v0.2.x`):**
  108. - **0 disk spills** across 134 consecutive `v0.2.x` MCP calls — strict `≤8.8 KB` UTF-8 inline cap prevents Antigravity's `output.txt` spill-to-disk extra turns.
  109. - **+75% effective throughput (`35.1 → 61.5 tok/s`)** and **`>60s` stall turns reduced from `31.2% → 7.4%`** in a same-session before/after `v0.2.1` comparison (99 vs. 374 turns) once multimodal screenshots were isolated out of `>180k`-token main sessions (`0` screenshots loaded in main coding sessions post-upgrade).
  110. - **0 lazy-load schema-read turns** — compact schemas are inlined directly in [`rules/AGENTS.md`](rules/AGENTS.md).
  111. <details>
  112. <summary>Methodology, quota sensitivity &amp; fixed prompt overhead</summary>
  113. - **Logical action normalization:** One `batch_search` with 5 queries counts as 5 actions, one `apply_edits` touching 6 hunks + `verify` counts as 7 actions, and one built-in `read_file`/`edit_file`/`run_command` counts as 1 action—ensuring apples-to-apples comparison per unit of work rather than raw tool-call counts.
  114. - **Quota weighting sensitivity:** Standard weighting (`1.0×` uncached input, `0.1×` cache read, `5.0×` output) yields **-31.5%** on main sessions (**-36.5%** on `≥150k` sessions). Results hold under conservative cache discounts (`1 : 0.25 : 4` → **-25.8%**) and high-output weighting (`1 : 0.1 : 8` → **-29.4%**).
  115. - **Fixed prompt overhead:** `rules/AGENTS.md` + `SessionStart` snapshot add **~1.7k tokens** once per session (cached from turn 2 onward, `~170` eq tokens/turn). Avoiding a single redundant turn (`~44k` eq tokens) or blind full-file read pays back this overhead **>20×**.
  116. </details>
  117. ### Verify on Your Own Transcripts (100% Offline)
  118. Measure your own Antigravity sessions locally — nothing ever leaves your machine:
  119. ```bash
  120. # Summarize all local Antigravity transcripts
  121. node dist/benchmark.mjs analyze --local
  122. # Analyze only the 10 most recently modified local transcripts
  123. node dist/benchmark.mjs analyze --local --recent 10
  124. # Slice sessions since install time and compare With vs. Without agy-fast
  125. node dist/benchmark.mjs analyze --local --since 2026-11-19T12:51:00+08:00 --split-usage
  126. # Compare two task runs (before vs. after)
  127. node dist/benchmark.mjs compare --before ./run-before/transcript.jsonl --after ./run-after/transcript.jsonl
  128. ```
  129. ## The 4 MCP Tools
  130. | Scenario | Preferred Tool | Why |
  131. | :--- | :--- | :--- |
  132. | Run ≥2 text, regex, glob, fuzzy, or read-only git queries | **`batch_search`** | 1 call for up to 20 concurrent queries (`<70 ms` local) |
  133. | Map unknown large files (`>300` lines) | **`read_slices`** | Symbol outlines across up to 40 files without full-file dumps |
  134. | Read small files (`<300` lines) or known line ranges | Built-in `read_file` | Locally allowlisted (`~7 ms`, `46 KB/call`, 0 remote check) |
  135. | Edit ≥2 hunks/files in one turn, or edit + verify | **`apply_edits`** | Atomic or `bestEffort` multi-file edits + optional inline `verify` |
  136. | Run ≥2 independent build, lint, test, or diagnostic commands | **`run_checks`** | Concurrent execution with compact error-focused output |
  137. ### 1. `batch_search` — Concurrent Code, Path & Read-Only Git Search
  138. - **Up to 20 mixed queries per call:** `text`, `regex`, `glob`, `fuzzy`, and read-only `git` (`status`, `diff`, `log`, `show`, `branch`, `rev-parse`, `ls-files`, `blame`, `shortlog`, `describe`, `tag --list`, `remote -v`), including cross-repo lookups via per-query `path`.
  139. - **Precision controls:** Filter by `fileType` (e.g., `typescript`, `python`, `gdscript`) or `glob`, switch `outputMode` (`content` | `files_with_matches` | `count`), enable cross-line `multiline: true` (where `count` reports occurrences per file), use `type: "fuzzy"` for case-insensitive path matching, or pass `sortBy: "mtime"` on `glob` queries for newest-first results.
  140. - **Resilient & inline-safe:** Uses `rg` when installed with automatic Node.js fallback, auto-recovers common LLM `glob` mistakes (`|` alternation and bare keywords), and enforces the `≤8.8 KB` UTF-8 cap.
  141. ### 2. `apply_edits` — Atomic Multi-Hunk & Multi-File Edits + Verify
  142. - **Up to 50 files per call:** Validates every hunk and overlap in memory before touching disk (`0` files modified on error), or pass `bestEffort: true` on larger batches so valid files commit atomically per file while failed files return closest-line / drifted-line snippets for targeted retry.
  143. - **Faithful file handling:** Preserves CRLF line endings, file permissions, and symlinks; supports `startLine`/`endLine`, `replaceAll`, `create`/`overwrite`, and `dryRun`.
  144. - **Inline `verify`:** Optionally runs a build/typecheck/test command right after writing, stripping noisy logs (e.g., Pino/Bunyan info JSON, Maven footers) so root causes stay within budget.
  145. ### 3. `read_slices` — Multi-Language Symbol Outlines & Targeted Slices
  146. - **Up to 40 file outlines per call** *(10–20 recommended per batch)*: Extracts exact `[startLine, endLine]` symbol/section maps for **TypeScript / JavaScript** *(incl. Fastify/Express routes & test blocks)*, **Vue / Svelte SFC**, **CSS / SCSS / Less**, **Python**, **Godot GDScript (`.gd`)**, **Go**, **Rust**, **Java / Kotlin / C# / Swift / C / C++**, **Shell**, **Markdown**, and **JSON / YAML**.
  147. - **Targeted slices:** Optionally piggyback tiny `[startLine, endLine]` slices in the same call under the `≤8.8 KB` UTF-8 budget, while steering bulk range reads (`>150` lines) to parallel built-in `read_file`.
  148. ### 4. `run_checks` — Concurrent Build, Lint, Test & Diagnostic Runner
  149. - **Up to 12 commands per call:** Executes independent non-interactive shell commands in a bounded worker pool (`concurrency: 1..8`) with per-command timeouts (`1..300 s`) and ANSI escape stripping.
  150. - **Adaptive output modes:** `compact` (default) collapses passing commands to 1-line summaries and extracts focused failure traces; `full` preserves stdout for diagnostic or audit scripts within the `≤8.8 KB` inline cap.
  151. <details>
  152. <summary>Example batched MCP calls (JSON payloads)</summary>
  153. **`batch_search` — 3 heterogeneous queries in one call:**
  154. ```json
  155. {
  156. "queries": [
  157. { "type": "regex", "pattern": "export (async )?function", "fileType": "typescript", "outputMode": "files_with_matches" },
  158. { "type": "fuzzy", "pattern": "mcp server" },
  159. { "type": "git", "pattern": "status --short" }
  160. ]
  161. }
  162. ```
  163. **`apply_edits` — Multi-file edit + inline typecheck in one call:**
  164. ```json
  165. {
  166. "edits": [
  167. { "path": "src/a.ts", "hunks": [{ "target": "oldCall()", "replacement": "newCall()" }] },
  168. { "path": "src/b.ts", "hunks": [{ "target": "v1", "replacement": "v2", "startLine": 10, "endLine": 25 }] }
  169. ],
  170. "bestEffort": true,
  171. "verify": "npm run typecheck"
  172. }
  173. ```
  174. </details>
  175. ## Updates, Diagnostics, and Uninstall
  176. ### Day-to-Day CLI Reference
  177. | Task | Command |
  178. | :--- | :--- |
  179. | **Health check** *(Node, `rg`, live MCP handshake, `SessionStart` latency, manifest validation)* | `node ~/.gemini/config/plugins/agy-fast/scripts/install.mjs doctor` |
  180. | **Check for updates** *(read-only)* | `node ~/.gemini/config/plugins/agy-fast/scripts/install.mjs update --check` |
  181. | **Update now** *(stages & runs MCP smoke test before swapping)* | `node ~/.gemini/config/plugins/agy-fast/scripts/install.mjs update` |
  182. | **Configure auto-update** *(`auto` \| `notify` \| `off`)* | `node ~/.gemini/config/plugins/agy-fast/scripts/install.mjs config --auto-update notify --interval 24` |
  183. | **Uninstall cleanly** | `node ~/.gemini/config/plugins/agy-fast/scripts/install.mjs uninstall` |
  184. > From a local repository clone, `./install.sh <command>` (macOS/Linux) and `.\install.ps1 <command>` (Windows) accept the same subcommands (`doctor`, `update`, `config`, `uninstall`).
  185. ### Zero-Latency Background Auto-Updater
  186. - **0 ms session startup cost:** Every **12 hours**, `SessionStart` spawns a detached background process (`dist/updater.mjs`) *after* emitting the workspace snapshot.
  187. - **Verified before swap:** Checks GitHub release tags (`git ls-remote --tags` with GitHub Releases API fallback), clones into an isolated staging directory, runs a **live 4-tool MCP smoke test**, and atomically replaces runtime files—automatically rejecting broken releases.
  188. - **Safe by design:** Standard copied installs and clean direct `git clone` installs auto-update automatically; `--link` dev symlinks and dirty git repos are **never overwritten** (notify-only). A one-time notice appears in the next new session when an update is applied or available.
  189. <details>
  190. <summary>Environment variable overrides &amp; applying updates</summary>
  191. | Environment Variable | Values / Default | Purpose |
  192. | :--- | :--- | :--- |
  193. | `AGY_FAST_AUTO_UPDATE` | `auto` *(default)* \| `notify` \| `off` | Overrides saved auto-update mode |
  194. | `AGY_FAST_UPDATE_INTERVAL_HOURS` | `12` *(default; `1..720`)* | Background check interval in hours |
  195. | `AGY_FAST_UPDATE_CHANNEL` | `release` *(default)* \| `main` | Track stable semver release tags or `main` |
  196. | `AGY_FAST_REPO` | Git URL | Use a custom mirror or fork |
  197. **Applying updates:** Updated `rules/AGENTS.md` and `dist/session-start.mjs` take effect automatically on the next new session. In Antigravity Desktop 2.0 / IDE, reload or restart the window once to refresh the resident MCP server process.
  198. </details>
  199. ## Policy Compliance and Safety
  200. - **Official interfaces only:** Built exclusively on documented Antigravity extension APIs (`plugin.json`, `mcp_config.json`, `hooks.json`, `rules/AGENTS.md`).
  201. - **Non-invasive by design:** Never alters or interferes with Antigravity's native UI, **diff viewer/window**, **modified-files list**, checkpoints, or built-in tools.
  202. - **No safety or binary tampering:** Does not patch client binaries, alter authentication, or bypass Antigravity's built-in safety evaluations.
  203. - **Zero per-step blocking hooks:** Registers no `PreToolUse`, `PostToolUse`, `PreInvocation`, `PostInvocation`, or `Stop` hooks.
  204. - **Fail-open startup hook:** On unreadable directories or malformed stdin, `SessionStart` returns `{}` and exits cleanly—never blocking a session.
  205. - **Local-first & auditable:** All 4 MCP tools and the benchmark CLI run 100% locally with zero code or transcript uploads; shipped `dist/*.mjs` bundles are unminified ESM so every executed line is directly auditable.
  206. ## FAQ
  207. ### Does `agy-fast` affect Antigravity's native UI, diff viewer, or modified-files list?
  208. No. `agy-fast` runs strictly as a standard local Stdio MCP plugin with a `SessionStart` hook. It never patches the Antigravity client or interferes with the native diff viewer/window, modified-files list, checkpoints, built-in tools, or safety checks.
  209. ### Does `agy-fast` save token quota and upload bandwidth in addition to reducing latency?
  210. Yes. Across 10 days of sanitized real coding transcripts, batching ~4.9 operations per MCP call reduced LLM turns by **33.7%**, estimated billable token quota by **31.5%** (**36.5%** in `≥150k`-token sessions; **54.0%** in subagents), tool-return upload bytes by **39.7%**, and `>60s` stall turns from **31.2% → 7.4%**.
  211. ### Does `agy-fast` replace Antigravity's built-in `read_file` or `edit_file` tools?
  212. No — it complements them in a hybrid workflow. Built-in `read_file` is locally allowlisted (`~7 ms`, `46 KB/call`, zero remote check), so `agy-fast` routes small files and known line ranges straight to parallel built-in `read_file`, while handling multi-query searches, symbol outlines, multi-hunk/multi-file edits with inline `verify`, and concurrent checks via MCP.
  213. ### Which Antigravity editions and operating systems are supported?
  214. `agy-fast` supports **Antigravity CLI (`agy`)**, **Antigravity Desktop 2.0**, and **Antigravity IDE** across **macOS, Linux, and Windows**. It requires **Node.js >= 20** with zero runtime dependencies (optional `ripgrep`/`rg` with automatic pure-Node.js fallback).
  215. ### Does `agy-fast` send any code or transcript data to external servers?
  216. No. All MCP tools, the `SessionStart` snapshot hook, and the transcript benchmark CLI operate 100% locally. The only optional network request is the background update checker checking public GitHub release tags once every 12 hours, which can be switched to `notify` or disabled completely via `config --auto-update off` or `AGY_FAST_AUTO_UPDATE=off`.
  217. ### How do I verify the speedup or uninstall `agy-fast` cleanly?
  218. Run `node dist/benchmark.mjs analyze --local --split-usage` to benchmark your own local transcripts completely offline. To remove `agy-fast` at any time, run `node ~/.gemini/config/plugins/agy-fast/scripts/install.mjs uninstall`.
  219. ## Development
  220. ```bash
  221. npm install
  222. npm run typecheck # Strict TypeScript check
  223. npm run build # Bundle src/*.ts -> dist/*.mjs with esbuild
  224. npm test # Run unit, MCP protocol, hook, benchmark, and installer tests
  225. npm run bench:stress # Run 12-scenario end-to-end stress & performance benchmark
  226. npm run validate # Run `agy plugin validate --strict .`
  227. ```
  228. ## Star History
  229. <a href="https://www.star-history.com/#INK-Q/agy-fast&Date">
  230. <picture>
  231. <source media="(prefers-color-scheme: dark)" srcset="https://api.star-history.com/svg?repos=INK-Q/agy-fast&type=Date&theme=dark" />
  232. <source media="(prefers-color-scheme: light)" srcset="https://api.star-history.com/svg?repos=INK-Q/agy-fast&type=Date" />
  233. <img alt="Star History Chart" src="https://api.star-history.com/svg?repos=INK-Q/agy-fast&type=Date" />
  234. </picture>
  235. </a>
  236. ## License
  237. [MIT](LICENSE)

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.