Claude Code × Antigravity + Codex + Copilot + Cursor + Grok + Kimi — MCP Bridge

Claude Code bridging Google Antigravity, OpenAI Codex, GitHub Copilot, and Cursor

Drive six external coding CLIs — Google's Antigravity (Gemini 3.6 Flash), OpenAI Codex, the GitHub Copilot CLI, Cursor, and the two experimental newcomers Grok Build and Kimi Code — as sub-agents inside Claude Code. Text answers, image generation, real repo work, and parallel swarms, on quota you already pay for.

One MCP server, six backends — four verified, two experimental. It exposes Google Antigravity, OpenAI Codex, the GitHub Copilot CLI, Cursor, and now xAI's Grok Build and Moonshot's Kimi Code to Claude Code as clean MCP tools so you can delegate work to a different model family mid-task — without leaving your terminal, and on the subscriptions you already have. Each backend is independent: install one, or all six.

🛰️ Antigravity (agy, Gemini 3.6 Flash High). Fast, cheap tool-calling — and the only backend with an image model. Its headless print mode (agy -p) historically had a stdout bug: it wrote the answer to the controlling terminal instead of its stdout, so anything capturing stdout got nothing (and, under a TUI, agy's text leaked into the host's prompt). agy 1.0.15 fixed this on Windows — -p now writes the clean answer to stdout — so the bridge prefers stdout and falls back to reading agy's own transcript files only when stdout is empty (older agy, non-Windows, or --sandbox runs). It still detaches agy from the terminal so older versions can't leak.
🤖 Codex (codex exec, OpenAI). A strong reasoner for real code/repo work. It writes its final message straight to a file the bridge asks for (no scraping), supports model selection, and has a real, enforced sandbox.
🐙 Copilot (copilot -p, GitHub). GitHub's agentic coder. Stdout-native like Codex (-s prints just the answer), with model selection (--model), a best-effort tool/path permission knob, and a deterministic resume mechanism (the bridge sets each session's UUID itself).
✳️ Cursor (cursor-agent -p, Cursor). Cursor's agentic coder, with the widest model menu — GPT, Claude, Grok, and Composer via --model (validated against cursor-agent models). Stdout-native like Codex/Copilot (--output-format text prints just the answer), an agent-enforced sandbox (read-only via --mode ask), and a deterministic resume mechanism (the bridge mints each chat's id itself via create-chat). No image model.
🧪 Grok Build (grok -p, xAI) — EXPERIMENTAL. xAI's terminal coding agent, and the only backend besides Codex with a real OS sandbox — though only on Linux/macOS. Stdout-native (--output-format json returns the answer and the session id), with -r resume, streaming-json for watch mode, and full swarm support. Never verified end-to-end — see below.
🌙 Kimi Code (kimi -p, Moonshot) — EXPERIMENTAL. Moonshot's terminal coding agent (Kimi K2 family). Stdout-native (--output-format text), resumes per working directory (-c). No sandbox — print mode auto-executes every tool, like Antigravity. Never verified end-to-end — see below.

They share the same niceties: a *_continue to resume a thread, a live "watch" window to see the agent work, a unified agent_swarm that runs many tasks in parallel across all backends at once, and *_status diagnostics that spend no quota. (Kimi is the one exception: no watch or swarm support yet — see Experimental backends.)

Important

Grok Build and Kimi Code ship unverified, and I need your help. I don't have a Grok or Kimi subscription, so no authenticated round-trip has ever run against either backend. Everything up to each CLI's auth wall is verified live — flag surface, error shapes, model list, on-disk layout — but everything behind it comes from vendor docs and could be wrong. If you have either subscription, one issue from the verification template is the single most useful contribution you can make. Even confirming one checkbox helps. Full detail →

Warning

This runs unsandboxed code with your privileges. agy -p auto-executes its tools (read/write files, run shell commands, reach the network) with no usable approval gate — its --sandbox blocks only shell commands, leaving file writes and network egress wide open. codex exec also runs autonomously, but its sandbox flag (default read-only) is a real, enforced boundary. copilot -p runs headless with --allow-all-tools; its sandbox maps to best-effort tool/path permissions (read-only denies the local write/shell tools) — safer than agy, but not an OS sandbox like Codex's. cursor-agent -p runs headless with --trust (and --force for writes); its sandbox is agent-enforced (read-only = --mode ask, which makes the write/shell tools unavailable) — best-effort like Copilot, not an OS sandbox. grok -p runs headless with --always-approve; its sandbox maps to a real OS profile (Landlock/Seatbelt) — but only on Linux and macOS, and on Windows grok silently continues without enforcement, so read-only there rests on an agent-enforced tool allowlist. kimi -p has no sandbox at all and auto-executes every tool, like agy. In all six cases the workspace argument is a starting context, not a security boundary. Only use these with trusted prompts on trusted content; for real isolation, run the bridge inside a container or VM. Full details →

Why you'd want this


🧠 Second opinion	Ask a different model family — Gemini or GPT — mid-task without switching tools.
🎨 Image generation	Have Gemini draw an image and get the saved file back — no extra API key or image tool.
🛠️ Real coding sub-agent	Hand a focused repo task to Codex with a real `workspace-write` sandbox.
💸 Cheap delegation	Burn Antigravity / Codex quota on grunt work instead of Claude tokens.
🐝 Parallel fan-out	Run N tasks at once, mixing Gemini and Codex workers in a single swarm.
📁 Cross-repo reads	Point a worker at another project directory and let it read/answer there.
🔌 Zero new auth	Piggybacks the logins you already did — no keys for the bridge to manage.

The backends at a glance

The bridge normalizes every CLI into the same shape, but they differ where it matters. Pick per task. The four verified backends first; the two experimental ones follow.

	🛰️ Antigravity (`agy`)	🤖 Codex (`codex exec`)	🐙 Copilot (`copilot -p`)	✳️ Cursor (`cursor-agent -p`)
Model	Selectable via `model` (agy's `--model`); Gemini 3.6 Flash (High) default (see Model & auth)	Selectable via `model` (codex's `-m`)	Selectable via `model` (`--model`)	Selectable via `model` (`--model`), validated against `cursor-agent models`
Best at	Fast, cheap tool-calling; quick answers	Heavier reasoning; real code/repo work	Agentic coding; real code/repo work	Agentic coding; wide model menu (GPT/Claude/Grok/Composer)
Image generation	✅ `antigravity_image` (+ `antigravity_image_swarm`)	❌ no image model	❌ no image model	❌ no image model
Sandbox	❌ no real boundary (`--sandbox` blocks only shell)	✅ real, enforced: `read-only` / `workspace-write` / `danger-full-access`	⚠️ best-effort: tool/path permissions (`read-only` denies write/shell) — not an OS sandbox	⚠️ agent-enforced: mode/force (`read-only` = `--mode ask`, write/shell tools unavailable) — not an OS sandbox
How the answer is read	`--output-format json` on agy 1.1.8+ (`stream-json` when watching); else stdout, else scraped from `transcript.jsonl`	Written to a file via `-o/--output-last-message`	stdout (`-s` silent mode)	stdout (`--output-format text`)
Continue mechanism	Pins the workspace's conversation id (`--conversation`)	Resumes the session id (`codex exec resume <id>`)	Resumes a self-set session UUID (`--session-id`)	Mints a chat id (`create-chat`) and resumes it (`--resume <id>`)
Auth	OS credential store (AI Pro session)	`codex login` (ChatGPT account or API key)	OS credential store (`copilot login`) or a GitHub token env	`cursor-agent login` (OS credential store) or `CURSOR_API_KEY`
In a swarm	Runs with an isolated `HOME` to avoid state races	Fresh one-shot — needs no isolation	Fresh one-shot — needs no isolation	Fresh one-shot — needs no isolation

🧪 The two experimental backends — and how you can help

Grok Build and Kimi Code are wired in exactly like the other four, with one honest difference: no authenticated round-trip has ever run against either. I don't have a SuperGrok / X Premium+ subscription or a Kimi plan, so I cannot prove they answer. They ship anyway because a bridge nobody can install is a bridge nobody can verify — and because the parts that usually rot are already pinned down.

What is verified live (each CLI installed, run, and observed — just never logged in):

	🧪 Grok Build (`grok -p`)	🌙 Kimi Code (`kimi -p`)
Verified against	grok 1.0.3 / Windows	kimi 0.29.1 / Windows
Flag surface	✅ read off the open-source clap definitions (xai-org/grok-build), then confirmed against live `grok --help`. Every argv the bridge can build was executed and parses cleanly	✅ confirmed against live `kimi --help`; also that `-p` rejects `--auto`/`--yolo` (print mode already self-approves, so the bridge passes neither)
Auth failure mode	✅ exit 1 + `{"type":"error","message":"Not signed in. …"}` on stdout; no browser, no hang	✅ exit 1 + stderr `No model configured`
Model list	✅ `grok models` answers while logged out — so auth checks and model validation cost nothing. Live default is `grok-4.5`, not the `grok-build` xAI's own docs still print	⚠️ none — Kimi has no `models` command; aliases are user-defined in `config.toml`, so `model` is a lenient pass-through
On-disk layout	✅ `~/.grok/` (`config.toml`, `auth.json`, `sessions/`, `logs/`); `GROK_HOME` really relocates it	✅ `~/.kimi-code/` (`config.toml`, `device_id`, `logs/`)
Concurrency	✅ parallel `grok -p` runs don't deadlock on `~/.grok`'s lock files	❔ untested

What is NOT verified — everything behind the auth wall:

the happy-path answer itself: Grok's json envelope (text / sessionId) and Kimi's stdout answer;
that -r / -c really restore context;
Grok's streaming-json event stream, which watch mode renders;
whether Grok's sandbox profiles behave as documented (and note: auth is checked before --sandbox and -m are validated, so a bad value can't even be observed while logged out — which is why the bridge validates both client-side).

Note

Deliberately scoped out for Kimi: agent_swarm and watch support. Both would depend on Kimi's stream-json envelope, and adding an unverified dependency on top of an unverified backend is how you get two bugs that mask each other. Grok gets both, because its stream format is documented in detail and its error events were observed live.

How to help

If you have either subscription, please open a verification issue. The template is a checklist — tick only what you actually saw. The first box ("a fresh ask returned a real answer") is worth more than all the others combined, and takes about a minute:

# 1. Does the setup look right? (spends no quota)
#    -> call grok_status / kimi_status from Claude Code
# 2. Does it answer?
#    -> call grok_ask("say hi") / kimi_ask("say hi")
# 3. If it fails, does the raw CLI fail the same way?
grok -p "say hi" --output-format json
kimi -p "say hi" --output-format text

That last command is the one I can't run from here, and it's what separates "the bridge is wrong" from "the CLI changed". Partial reports are welcome; so is a plain "it didn't work, here's the error".

How it works

All six backends run headless and one-shot per call; the bridge's job is to get a clean answer out of each and hand it to Claude Code as a plain string.

flowchart LR
    A([Claude Code]) -- "MCP tool call" --> B["bridge<br/>(server.py)"]
    B -- "antigravity_*" --> C[agy -p]
    B -- "codex_*" --> D[codex exec]
    B -- "copilot_*" --> E[copilot -p]
    B -- "cursor_*" --> F[cursor-agent -p]
    C -- "json / stream-json (1.1.8+)<br/>else stdout or transcript.jsonl / .db" --> B
    D -- "output-last-message file" --> B
    E -- "stdout (-s silent)" --> B
    F -- "stdout (--output-format text)" --> B
    B -- "plain text" --> A

Antigravity. On agy 1.1.8+ the bridge asks for structured output and reads a contractual field instead of guessing: plain calls use --output-format json and return its response, while watch mode uses --output-format stream-json and rebuilds the answer from the stream's terminal result event (the same shape the Cursor bridge already used). Both also carry a conversation_id, which the bridge records so antigravity_continue pins exactly the thread it last ran in that workspace.

Older agy has no such flag, so the original path stays: on 1.0.15+ (Windows) agy -p writes its clean answer to stdout and the bridge returns that; on older agy — or non-Windows, or a --sandbox run — stdout is empty and the bridge falls back to agy's own transcript at:

~/.gemini/antigravity-cli/brain/<conv-id>/.system_generated/logs/transcript.jsonl

For that fallback it locates the conversation via cache/last_conversations.json (falling back to the newest brain/ directory touched since launch), streams the transcript, and returns the final source=MODEL, status=DONE, type=PLANNER_RESPONSE entry — the answer, minus the intermediate tool-calling steps (or the SQLite .db agy dual-writes, when no JSONL exists). This fallback still runs on 1.1.8+ whenever a run yields no result, so nothing depends on the structured path alone.

Codex. codex exec is well-behaved: the bridge passes -o/--output-last-message <file> and codex writes its final message straight there — no scraping. Continue works by capturing the session id from codex's own rollout files (~/.codex/sessions/.../rollout-*.jsonl) and resuming with codex exec resume <id>, falling back to the newest on-disk session for that cwd after a server restart.

Copilot. copilot -p "<prompt>" -s runs a prompt non-interactively and prints the clean final answer to stdout — the bridge reads it there, no scraping. It runs headless with --allow-all-tools --no-ask-user --no-auto-update (so it never blocks on a prompt), and disables copilot's flaky builtin GitHub-API MCP by default for predictable latency (COPILOT_GITHUB_MCP=1 re-enables it). Continue is deterministic: copilot's --session-id <uuid> both sets a new session's id and resumes an existing one, so the bridge generates the UUID itself, pins it to the workspace, and resumes that exact session — falling back after a restart to the newest on-disk session (~/.copilot/session-state/<id>/workspace.yaml) whose recorded cwd matches.

Cursor. cursor-agent -p --output-format text --trust "<prompt>" runs a prompt non-interactively and writes the clean final answer straight to stdout — the bridge reads it there, no scraping (--trust trusts the workspace so it never blocks on a prompt). Continue is deterministic and race-free: cursor-agent create-chat mints a fresh chat and prints its id, so the bridge mints the id itself, pins it to the workspace, and resumes that exact chat with -p --resume <chatId> — no rollout-scraping. After a restart it falls back to the newest on-disk chat under ~/.cursor/chats/<md5(workspace)>/<chat-id>/ whose meta.json cwd matches (the chat-dir hash is itself md5 of the workspace path).

Set up in 60 seconds

Prerequisites — install whichever backend(s) you want, and sign in once each:

Antigravity: install agy and sign in to Antigravity once (via the IDE or agy -i).
Codex: install codex and run codex login once (ChatGPT account or API key).
Copilot: install copilot (npm i -g @github/copilot, or winget install GitHub.Copilot) and run copilot then /login once (or set a COPILOT_GITHUB_TOKEN/GH_TOKEN env var).
Cursor: install cursor-agent (curl https://cursor.com/install -fsSL | bash) and run cursor-agent login once (or set a CURSOR_API_KEY env var).

You don't need all six — the tools for a missing CLI simply report "not found" via their *_status tool.

Recommended — no clone, you control updates

With uv installed, register the bridge straight from PyPI under mcpServers in ~/.claude.json — no path to hardcode, no git pull to remember:

"agent-intern": {
  "command": "uvx",
  "args": ["agent-intern"]
}

uvx pins to the version it first caches and does not auto-upgrade, so you never run an update you didn't choose — important, since the bridge runs unsandboxed code: a surprise (or compromised) release can't execute until you opt in. When the startup check warns that a newer release is out, upgrade deliberately and restart Claude Code:

uvx agent-intern@latest      # fetch + run the newest release (refreshes uv's cache)

Tip

Prefer hands-off auto-updates? Put "args": ["agent-intern@latest"] in the config instead — every launch runs the newest release. Convenient, but it pulls new code without asking each time.

From source

Clone it instead if you want to hack on the bridge or pin a local copy:

git clone https://github.com/SinanTufekci/agent-intern.git
cd agent-intern
pip install fastmcp
python test_smoke.py        # 4 real round-trips (ask, continue, image, swarm) — prints four PASS lines

Note

The smoke test costs a tiny bit of quota and takes ~30–60 s. It exercises the Antigravity path.

Then point Claude Code at the absolute path to server.py under mcpServers in ~/.claude.json:

Windows	macOS / Linux
"agent-intern": { "command": "python", "args": ["C:\\path\\to\\server.py"] }	"agent-intern": { "command": "python3", "args": ["/path/to/server.py"] }

Restart Claude Code. Twenty-one tools appear, each prefixed mcp__agent-intern__:

Antigravity (5): antigravity_ask, antigravity_continue, antigravity_image, antigravity_image_swarm, antigravity_status
Codex (3): codex_ask, codex_continue, codex_status
Copilot (3): copilot_ask, copilot_continue, copilot_status
Cursor (3): cursor_ask, cursor_continue, cursor_status
Grok (3, experimental): grok_ask, grok_continue, grok_status
Kimi (3, experimental): kimi_ask, kimi_continue, kimi_status
Shared (1): agent_swarm — fans a list of tasks out across five backends in one run (everything but Kimi)

The single-prompt tools — Antigravity, Codex, Copilot, Cursor, and Grok — take a watch=true flag for the live browser view (Watch mode). Kimi has no watch mode yet.

Note

Your client learns how to use the bridge on its own. The server ships MCP instructions — a short routing guide (when to reach for each tool, which backend to pick, and to pass workspace so the sub-agent has repo context) that a client like Claude Code injects into the model's context on connect, as an "MCP Server Instructions" block. So the host model knows how and when to drive these tools without you explaining them — you can just ask for the result.

"Use antigravity_ask to summarize the README of this repo in three bullets." → Claude routes the prompt through the bridge, agy reads the file under the workspace root, and the answer comes back as a plain string. Swap in codex_ask, copilot_ask, or cursor_ask to have GPT, Copilot, or Cursor do the same.

Tools

🛰️ Antigravity

Tool	Purpose
`antigravity_ask(prompt, workspace?, model?, timeout_s?=180, watch?=false)`	Start a new Antigravity conversation. `model` selects the model (agy's `--model`, e.g. `"claude-sonnet-4-6"`); validated against `agy models`, defaults to your `settings.json` model. `watch=true` opens the live browser view (Watch mode).
`antigravity_continue(prompt, workspace?, model?, timeout_s?=180, watch?=false)`	Continue the conversation rooted at `workspace` (pinned by id). agy's model is per-invocation, so `model` can differ from the original ask. `watch=true` opens the live view.
`antigravity_image(prompt, output_path?, workspace?, timeout_s?=240, watch?=false)`	Generate an image; saves the file (extension corrected to the real bytes) and returns its path + format/size. `watch=true` streams progress and shows the image inline.
`antigravity_image_swarm(prompts, output_paths?, workspaces?, max_concurrency?=4, timeout_s?=240, watch?=false)`	Generate several images in parallel (one worker per prompt).
`antigravity_status()`	Setup diagnostics: the bridge's own version + whether a newer release is available, remaining AI Pro quota per model family (agy 1.1.11+), plus agy version/compat, state dirs, and newest-transcript readability. Spends no quota.

🤖 Codex

Tool	Purpose
`codex_ask(prompt, workspace?, sandbox?="read-only", model?, timeout_s?=180, watch?=false)`	Start a new Codex session. `sandbox` is a real boundary (see Codex bridge); `model` selects the model (`-m`). `watch=true` opens the live view, streaming codex's steps from its `--json` event stream.
`codex_continue(prompt, workspace?, timeout_s?=180, watch?=false)`	Continue the Codex session rooted at `workspace` — resumes the exact session id, falling back to the newest on-disk session for that cwd after a server restart. The resumed session keeps its original sandbox and model. `watch=true` opens the live view.
`codex_status()`	Setup diagnostics: codex version, login status (`codex login status`), sessions dir. Spends no quota.

🐙 Copilot

Tool	Purpose
`copilot_ask(prompt, workspace?, sandbox?="read-only", model?, timeout_s?=180, watch?=false)`	Start a new Copilot session. `sandbox` maps to copilot's tool/path permissions (best-effort, not an OS sandbox — see Copilot bridge); `model` selects the model (`--model`). `watch=true` opens the live view, streaming copilot's steps from its `--output-format json` event stream.
`copilot_continue(prompt, workspace?, sandbox?="read-only", timeout_s?=180, watch?=false)`	Continue the Copilot session rooted at `workspace` — resumes the exact self-set session id, falling back to the newest on-disk session for that cwd after a restart. Unlike Codex, `sandbox` applies here too (copilot re-applies permissions each turn). `watch=true` opens the live view.
`copilot_status()`	Setup diagnostics: copilot version, an auth hint (no `login status` command exists, so best-effort), session-state dir. Spends no quota.

✳️ Cursor

Tool	Purpose
`cursor_ask(prompt, workspace?, sandbox?="read-only", model?, timeout_s?=180, watch?=false)`	Start a new Cursor chat. `sandbox` maps to cursor's mode/force flags (agent-enforced, not an OS sandbox — see Cursor bridge); `model` selects the model (`--model`, validated against `cursor-agent models`). `watch=true` opens the live view, streaming cursor's steps from its `--output-format stream-json` event stream.
`cursor_continue(prompt, workspace?, sandbox?="read-only", timeout_s?=180, watch?=false)`	Continue the Cursor chat rooted at `workspace` — resumes the exact chat id the bridge minted (`create-chat` + `--resume`), falling back to the newest on-disk chat for that cwd after a restart. `watch=true` opens the live view.
`cursor_status()`	Setup diagnostics: the bridge's own version + whether a newer release is available, plus cursor version and login status (`cursor-agent status`). Spends no quota.

🧪 Grok Build (experimental — unverified)

Tool	Purpose
`grok_ask(prompt, workspace?, sandbox?="read-only", model?, timeout_s?=180, watch?=false)`	Start a new Grok session. `sandbox` maps to grok's `--sandbox` profile plus a tool allowlist — a real OS boundary on Linux/macOS only (see Grok bridge); `model` selects the model (`-m`, validated against `grok models`). `watch=true` opens the live view, streaming grok's steps from its `--output-format streaming-json` event stream.
`grok_continue(prompt, workspace?, sandbox?="read-only", timeout_s?=180, watch?=false)`	Continue the Grok session rooted at `workspace` — resumes the exact session id grok returned (`-r`), falling back to grok's own "most recent session for this cwd" (`-c`) after a restart. `sandbox` applies here too. `watch=true` opens the live view.
`grok_status()`	Setup diagnostics: the bridge's own version + whether a newer release is available, plus grok version, auth state, and the model list — the last two both from `grok models`, which answers even while logged out. Spends no quota.

🌙 Kimi Code (experimental — unverified)

Tool	Purpose
`kimi_ask(prompt, workspace?, model?, timeout_s?=180)`	Start a new Kimi session. No `sandbox` argument — Kimi print mode has no sandbox and auto-executes every tool. `model` is a lenient pass-through (`-m`, an alias from your `config.toml`); Kimi has no model list to validate against. No watch mode.
`kimi_continue(prompt, workspace?, timeout_s?=180)`	Continue the Kimi session rooted at `workspace` (`-c`). Kimi scopes sessions per working directory, so there's no id to track — and no restart problem either.
`kimi_status()`	Setup diagnostics: bridge version + update check, kimi version, whether a provider is configured (`kimi provider list` — the auth proxy), and the data dir. Spends no quota.

🐝 Shared

Tool	Purpose
`agent_swarm(tasks, max_concurrency?=4, timeout_s?=180, watch?=false)`	Run several tasks in parallel across five backends — each task names its `backend` (`antigravity`, `codex`, `copilot`, `cursor`, or `grok`) plus a `prompt` (an optional `model` for any backend, and `sandbox` for Codex/Copilot/Cursor/Grok). Every answer comes back in one block; `watch=true` opens the live dashboard (Swarm). Kimi is not available here — see Experimental backends.

workspace defaults to the MCP server's current working directory. Point it at a real project dir for context-aware answers — every backend gives the model access to files under that root (Codex, Copilot, and Cursor honoring their sandbox).

antigravity_image forces agy to save to an explicit absolute path — without one, agy falls back to its own scratch dir (~/.gemini/antigravity-cli/scratch/). It then corrects the file extension to match the real bytes: agy's image model picks the format itself (JPEG for photo-like images, PNG for flat graphics), so a requested out.png may come back as out.jpg. The returned path always reflects the true format.

🤖 Codex bridge — the well-behaved sibling

codex exec writes its final message to a file the bridge asks for via -o/--output-last-message, so the answer comes back without any scraping (where agy needed a transcript workaround before 1.0.15 fixed its stdout). Three things make Codex worth reaching for over Antigravity:

Real sandbox. sandbox accepts read-only (default — reads and answers, writes nothing), workspace-write (may edit files under the workspace), or danger-full-access (no sandbox — avoid). Unlike agy's no-op --sandbox, codex's -s actually enforces this. codex exec has no interactive approval gate, so this flag is your safety boundary — opt into write access deliberately.
Model selection works. model maps to codex's -m. (agy's --model works in print mode too as of 1.0.16; every backend now exposes the same model knob, except Kimi, which has no list to validate against.)
Stronger reasoning. Codex is a coding agent, not an image model — there's no codex_image. Its strength is reasoning and real code/repo work; hand it the jobs that need a heavier model.

Auth. Uses your existing Codex login (ChatGPT account or API key). Run codex login once; check with codex_status. No new keys for the bridge to manage.

Warning

codex exec runs the model as an autonomous agent with no interactive approval gate. The sandbox flag (default read-only) is the real boundary, but workspace-write / danger-full-access let it modify files — and a swarm runs N agents at once. Only use it with trusted prompts on trusted content.

🐙 Copilot bridge — GitHub's agentic coder

The GitHub Copilot CLI (copilot, from @github/copilot) is stdout-native like Codex: copilot -p "<prompt>" -s runs a prompt non-interactively and prints just the final answer to stdout, so the bridge reads it there — no scraping. What makes it worth reaching for:

Model selection. model maps to copilot's --model; auto lets Copilot pick. Unlike the agy and cursor tools, the bridge can't validate this — copilot exposes no non-interactive model list — and the working set is account-dependent: on a Copilot Pro account here, auto worked while gpt-5.3-codex, claude-sonnet-4.6, and even GitHub's own --help example gpt-5.4 were all rejected as "not available". So omit model (account default) or pass auto unless you know your plan's ids; an unavailable one errors immediately with copilot's message, costing a call.
Deterministic, race-free continue. copilot's --session-id <uuid> both sets a new session's id and resumes an existing one, so the bridge generates the UUID itself and pins it to the workspace — no rollout-scraping. After a restart it falls back to the newest on-disk session (~/.copilot/session-state/<id>/workspace.yaml) whose recorded cwd matches.
Fast by default. Runs with --allow-all-tools --no-ask-user --no-auto-update, and disables copilot's builtin GitHub-API MCP (--disable-builtin-mcps) because its flaky HTTP connect can stall a call up to ~60 s. Set COPILOT_GITHUB_MCP=1 to keep it (for Copilot's issue/PR/repo tools).

Sandbox is best-effort, not enforced. Unlike Codex's OS sandbox, copilot's boundary is tool/path permissions. The sandbox knob maps to copilot flags for a uniform cross-backend field:

read-only (default) — auto-approves tools so it runs headless, then denies the local write and shell tools (--deny-tool). Best-effort: it is not an OS sandbox, and network/MCP tools can still act. For a hard read-only boundary, use codex_ask instead.
workspace-write — writes allowed, but file access stays confined to the workspace (no --allow-all-paths).
danger-full-access — --allow-all (tools + all paths + all URLs). Avoid.

Auth. Uses your existing Copilot login — run copilot then /login once (stored in the OS credential store), or set COPILOT_GITHUB_TOKEN/GH_TOKEN/GITHUB_TOKEN for headless use. Check with copilot_status. If copilot isn't on PATH (the winget install can land off a stale PATH), set COPILOT_BIN to its full path — e.g. %LOCALAPPDATA%\Microsoft\WinGet\Packages\GitHub.Copilot_*\copilot.exe.

Warning

copilot -p runs the model as an autonomous agent with --allow-all-tools (required to run headless). Its sandbox is best-effort tool/path permissions, not an OS sandbox — safer than agy, weaker than Codex's read-only. Only use it with trusted prompts on trusted content.

✳️ Cursor bridge — the widest model menu

Cursor's agent CLI (cursor-agent, from cursor.com/cli) is stdout-native like Codex and Copilot: cursor-agent -p --output-format text --trust "<prompt>" runs a prompt non-interactively and writes just the final answer to stdout, so the bridge reads it there — no scraping (--trust trusts the workspace so it won't block on a prompt). What makes it worth reaching for:

The widest model menu. model maps to cursor's --model (e.g. auto, gpt-5.2, claude-opus-4-8-high, composer-2.5, cursor-grok-4.5-high) — GPT, Claude, Grok, and Composer in one place, ~190 ids at the time of writing. cursor bakes the effort and speed axes into the id (…-low / -high / -xhigh / -max, each with a -fast twin), and also accepts a bracket form on the family base, e.g. claude-opus-4-8[context=1m,effort=high]. The bridge validates against cursor-agent models and rejects a typo up front (like agy), accepting either an exact id or a family base. Omit model to use your Cursor account default. cursor reshuffles this list often — run cursor-agent models (or cursor_status) rather than trusting an example here.
Deterministic, race-free continue. cursor-agent create-chat mints a fresh chat and prints its id, and -p --resume <chatId> resumes that exact chat — so the bridge mints the id itself, pins it to the workspace, and resumes deterministically (no rollout-scraping, same idea as Copilot's self-set session id). After a restart it falls back to the newest on-disk chat under ~/.cursor/chats/<md5(workspace)>/<chat-id>/ whose meta.json cwd matches (the chat-dir hash is itself md5 of the workspace path).

Sandbox is agent-enforced, not an OS sandbox. Like Copilot, cursor's boundary is which tools the agent can reach, not an OS jail. The sandbox knob maps to cursor's mode/force flags for a uniform cross-backend field:

read-only (default) — --mode ask: the write and shell tools are unavailable, so cursor analyzes and answers but makes no edits (verified: it refuses to write files). Agent-enforced and best-effort — it is not an OS sandbox. For a hard read-only boundary, use codex_ask instead.
workspace-write — --force: edits and commands allowed, file access rooted at --workspace.
danger-full-access — --force --sandbox disabled (OS sandbox off). Avoid.

(Cursor also exposes an OS-level --sandbox enabled/disabled; the bridge drives the uniform field via mode/force.)

Auth. Uses your existing Cursor login — run cursor-agent login once (OS credential store), or set CURSOR_API_KEY for headless use. Check with cursor_status. If cursor-agent isn't reliably on PATH (the installer drops a cursor-agent.CMD shim a bare name can't launch on Windows), set CURSOR_BIN to its full path — mirrors the AGY_BIN/CODEX_BIN/COPILOT_BIN overrides.

Warning

cursor-agent -p runs the model as an autonomous agent with --trust (and --force when writes are allowed). Its sandbox is agent-enforced (read-only makes the write/shell tools unavailable), not an OS sandbox — safer than agy, weaker than Codex's read-only. Only use it with trusted prompts on trusted content.

🧪 Grok Build bridge — a real sandbox, on two of three platforms

Warning

EXPERIMENTAL — never verified end-to-end. Everything below the "Auth" line is confirmed against a live grok 1.0.3; the answer path is not. See Experimental backends, and please report what you find.

xAI's Grok Build (grok, installed with curl -fsSL https://x.ai/cli/install.sh | bash, or irm https://x.ai/cli/install.ps1 | iex on Windows) is stdout-native like Codex/Copilot/Cursor: grok -p "<prompt>" --output-format json runs a prompt non-interactively and writes a single JSON result object to stdout. What makes it interesting:

It's open source. xai-org/grok-build publishes the actual CLI source, so this bridge's flag surface was read off the real clap definitions rather than inferred from docs — then confirmed against grok --help. That's a much stronger footing than a docs-derived bridge, and it caught a live discrepancy: xAI's own headless docs use -m grok-build in their examples, but the real default on 1.0.3 is grok-4.5.
The answer carries its own session id. --output-format json returns {"text": …, "sessionId": …, "usage": …}, so the bridge pins that id and resumes the exact session with -r <id> — no id-minting dance like Cursor's, no rollout-scraping like Codex's. After a restart it falls back to -c, grok's own "most recent session for this cwd", so continue survives without ever reading grok's opaque SQLite session store.
Free auth + model checks. grok models answers while logged out (exit 0, printing You are not authenticated. and the catalogue), so grok_status and model validation cost nothing and need no login.

Sandbox is real — on Linux and macOS. This is the only backend besides Codex with an OS-enforced boundary, but read the platform caveat:

read-only (default) — --sandbox read-only plus a --tools allowlist (read_file,list_dir,grep,glob,web_search,web_fetch) plus --no-subagents.
workspace-write — --sandbox workspace: writes land in the workspace, ~/.grok, and temp.
danger-full-access — --sandbox off. Avoid.

Caution

On Windows, grok's OS sandbox does not apply. It's implemented with Landlock (Linux) and Seatbelt (macOS); where it can't be applied, xAI's docs say grok "logs a warning and continues without enforcement." That's why read-only here doesn't lean on the profile alone — the tool allowlist is agent-enforced and holds on every platform. An allowlist, not a denylist, precisely because it fails safe: a future grok that adds a new write tool can't silently slip through it. Note that MCP meta-tools stay available under an allowlist, so a configured MCP server could still write. For a hard boundary on every platform, use codex_ask.

Every mode also passes --always-approve: grok's headless mode does not auto-approve on its own (unlike agy and Kimi), and there's no human to answer a prompt. Containment comes from the profile and the allowlist, not from the approval gate.

Auth. grok login (browser OAuth), grok login --device-code (headless), or an XAI_API_KEY env var; credentials cache in ~/.grok/auth.json. Needs a SuperGrok or X Premium+ subscription. Check with grok_status. Set GROK_BIN to override the executable path — though the bridge already falls back to the installer's own ~/.grok/bin when grok isn't on PATH, which matters because the installer appends to the user PATH and that never reaches an already-running server process. GROK_HOME relocates the whole data dir. The bridge disables grok's background auto-updater per call via GROK_DISABLE_AUTOUPDATER=1 — a CLI that updates itself mid-session has broken this project before.

🌙 Kimi Code bridge — no sandbox, per-directory sessions

Warning

EXPERIMENTAL — never verified end-to-end. See Experimental backends.

Moonshot's Kimi Code (kimi, npm @moonshot-ai/kimi-code) runs the Kimi K2 family. kimi -p "<prompt>" --output-format text writes the clean final answer to stdout.

Continue is per-directory. Kimi scopes sessions to the working directory and exposes -c/--continue, so the bridge just re-runs with cwd=workspace and -c — no id to capture, and no restart problem. (-S/--session <id> exists but is deliberately unused: its on-disk format couldn't be verified.)
No model validation. Kimi has no models command; aliases are user-defined in ~/.kimi-code/config.toml under [models."<alias>"]. model is a lenient pass-through, so a bad alias surfaces as Kimi's own run-time error.
-p refuses --auto and --yolo (verified live on 0.29.1: "Cannot combine --prompt with …") because print mode is already self-approving — so the bridge passes neither.

Caution

Kimi has no sandbox and no sandbox argument. Print mode auto-executes every tool call with no approval gate — the same posture as agy's print mode. No flag makes it safe. Only use it with trusted prompts on trusted content.

Auth. kimi login (device-code OAuth) or an API key in ~/.kimi-code/config.toml (it does not read a bare env var). Check with kimi_status, which reads kimi provider list as the auth proxy. Set KIMI_BIN to override the executable path; KIMI_CODE_HOME relocates the data dir.

No swarm or watch support, deliberately — both would depend on Kimi's stream-json envelope, which no one has confirmed. They'll follow a successful verification report.

👁️ Watch mode — Agent Intern (experimental)

Pass watch=true to any single-prompt tool — antigravity_ask, antigravity_continue, antigravity_image, codex_ask, codex_continue, copilot_ask, copilot_continue, cursor_ask, cursor_continue, grok_ask, or grok_continue — to watch the agent work live in a little chat-style browser window called Agent Intern. The agent still runs headless; alongside it the bridge serves a tiny page on 127.0.0.1 and opens it in a small, chromeless app window that renders the exchange as a conversation: your prompt shows as a chat bubble, the agent's live steps stream in a collapsible "thinking" trace — its planner narration (▸), the real commands it runs ($), and completions (✓), read live (from agy's --output-format stream-json on 1.1.8+ — its transcript on older agy — or codex's / copilot's JSON event stream, or cursor's / grok's streaming-json) — and the final answer arrives as a Markdown card (and, for antigravity_image with watch=true, the generated image shown inline). A *_continue run opens with the prior turns of the conversation shown as history, so it reads as one ongoing thread rather than a blank new window. (A watched cursor_continue is the exception — Cursor stores its transcript in an opaque SQLite blob, so its window opens without visible prior-turn history.)

text ask / continue (agy, codex, copilot, or cursor)	`antigravity_image` — image inline

_{Real captures — the agent runs headless while the Agent Intern window renders the exchange as a chat conversation: your prompt as a CLAUDE bubble, live steps (▸ narration · $ commands · ✓ completions) in a collapsible trace, then the final Markdown answer or inline image.}

Cross-platform & best-effort. Prefers a Chromium browser (--app mode) for the windowed look; falls back to a normal browser window. If nothing can open, the run still completes and returns normally.
Window size. Set AGY_WATCH_WINDOW_SIZE (e.g. AGY_WATCH_WINDOW_SIZE=480,700) to resize the window; default is 560,760. Press Enter / Esc in the window to close it.
One window, reused — but concurrent runs stay separate. Repeated sequential watch calls reuse the already-open window instead of stacking a new one (the open page resets itself for the new run; the swarm dashboard rebuilds for the new fan-out). A run that starts while another watched run is still working gets its own window instead — so two concurrent single-worker runs (e.g. a codex_ask and a copilot_ask at once) each stream into their own view and never clobber each other. If you closed the window, the next run opens a fresh one. Set AGY_WATCH_ALWAYS_NEW=1 to force a new window every time.
Access control. The viewer is an HTTP server, and it serves your prompts, the answers, and the real commands the agents ran — so it binds 127.0.0.1 on an ephemeral port and requires two things on every request: a loopback Host header (which is what makes DNS rebinding fail — a rebound page arrives under the attacker's hostname) and a per-process token carried in the URL, which stops another local process or another user on a shared machine from simply connecting. The bridge puts the token in every URL it opens, so none of this is visible in normal use. Worth knowing because the server starts lazily but is never stopped: one watch=true run leaves the port listening for the life of the MCP server.
Chat layout & history. Prompts render as chat bubbles (labelled CLAUDE, since the MCP client writes them) — long ones clamp to a few lines with a show more / show less toggle — and answers as Markdown cards tagged with the backend (AGY / CODEX / COPILOT / CURSOR). A *_continue run seeds the window with the conversation's prior turns, read from each backend's own session store (agy's transcript, codex's rollout, copilot's events.jsonl; Cursor's store is opaque, so a watched cursor_continue opens without visible history). The swarm's per-worker detail window uses the same chat design for its one task.
Progress, keyboard & copy. Each panel shows a time progress bar (elapsed / timeout). The swarm dashboard adds an overall done/total bar and per-row time bars; use ↑/↓ to select a worker and ↵ to open its detail window. Answers render as Markdown with a copy button, and a "jump to latest" badge appears if you scroll up.
Coarse, not token-level. The backends flush their step stream in chunks, so you get a handful of live steps, not character streaming. The returned value is identical to the non-watch call. Nothing is sent anywhere but your own machine.

🐝 Swarm — run agents in parallel

agent_swarm fans a list of tasks out to workers that run truly concurrently (capped at max_concurrency, default 4), then returns every worker's result in one block. Each task names its own backend, so a single swarm can mix Antigravity (Gemini), Codex, Copilot, and Cursor workers — hand the reasoning-heavy jobs to Codex, Copilot, or Cursor and the quick ones to Gemini, all at once. Good for independent sub-tasks: summarise N files, ask the same question about N repos, fix N bugs. (antigravity_image_swarm stays separate — it generates N images, and only agy has an image model.)

agent_swarm(tasks=[
  {"backend": "antigravity", "prompt": "Summarise src/auth.py in 2 bullets."},
  {"backend": "codex", "prompt": "Find and fix the failing test in tests/",
   "sandbox": "workspace-write", "workspace": "./repo"},
  {"backend": "copilot", "prompt": "Explain what src/api.py exposes.",
   "sandbox": "read-only", "workspace": "./repo"},
  {"backend": "cursor", "prompt": "Draft a docstring for src/utils.py.",
   "model": "auto", "workspace": "./repo"},
  {"backend": "grok", "prompt": "List the public exports of src/index.ts.",
   "sandbox": "read-only", "model": "grok-4.5", "workspace": "./repo"},
])

Agent Swarm dashboard: workers running in parallel, each row showing its backend badge, repo, prompt, latest step and a per-worker time bar, while the overall done/total counter climbs

_{agent_swarm(..., watch=true) — one row per worker (with a backend badge); the done/total bar climbs as workers finish. Click a row (or ↑/↓ then ↵) to pop that agent into its own window.}

How it stays correct under concurrency. The single-agent agy tools serialize through a lock because agy rewrites last_conversations.json on every call, so concurrent runs sharing one state dir would race. The swarm sidesteps this: each agy worker runs with its own isolated HOME/USERPROFILE, so agy's brain/, cache/, and last_conversations.json never collide — no lock needed. Auth still works because agy reads it from the OS credential store, not from ~/.gemini (verified on agy 1.0.9). Codex, Copilot, and Cursor workers need no such isolation — each is a fresh one-shot (codex exec with its own -o file; copilot -p with its own self-set session id; cursor-agent -p with its own minted chat id). Each worker's cwd is its real workspace, so file access is unchanged. Measured ~2.8× speedup at 3 agy workers (the AI Pro backend does not serialize per-account); higher max_concurrency trades quota/rate-limit pressure for wall-clock.

Per-task fields — backend (antigravity/codex/copilot/cursor/grok) and prompt are required; workspace defaults to the server cwd; sandbox and model apply to Codex, Copilot, and Cursor (ignored for Antigravity). Swarm workers are one-shot — there is no *_continue for a swarm worker's session.
Error isolation — a worker that fails is reported in place; the others still return.
watch=true — opens a thin live Agent Swarm dashboard (one row per worker, with a backend badge, repo, prompt, and latest step). Click a row to pop that agent into its own window streaming its full step log.

Warning

A swarm launches N unsandboxed agents at once — N× the prompt-injection "lethal trifecta" surface of a single call (see Security). Only use it with trusted prompts on trusted content. Codex workers honor their enforced sandbox; Copilot and Cursor workers honor their best-effort sandbox; Antigravity workers have no real boundary.

Model & auth

	🛰️ Antigravity	🤖 Codex	🐙 Copilot	✳️ Cursor
Model	Selectable via the `model` argument (agy's `--model`, e.g. `"gemini-3.1-pro-high"`, `"claude-sonnet-4-6"`); omit to use the `"model"` field in agy's `settings.json` (`gemini-3.6-flash-high` by default as of 1.1.6). agy 1.1.5 replaced the old human labels with these slugs — the old `"Gemini 3.1 Pro (High)"` form no longer works. Switching model in `-p` used to hang (through ~1.0.14) but is fixed as of 1.0.16. An unknown model was silently ignored through 1.1.1 and hard-fails in `-p` as of 1.1.2; either way the bridge validates it against `agy models` and rejects a typo up front. Flash High is speed-optimized for cheap tool-calling; pick a bigger model for heavier work.	Selectable via the `model` argument (codex's `-m`). codex does not hang on a switch, so model choice is a first-class knob.	Selectable via the `model` argument (`--model`, e.g. `gpt-5.3-codex`, `claude-sonnet-4.6`, `auto`); omit for your account default. An unavailable model errors immediately.	Selectable via the `model` argument (`--model`, e.g. `gpt-5.2`, `claude-4-sonnet-thinking`, `auto`, or parameterized ids like `claude-opus-4-8[context=1m]`); a wide GPT/Claude/Grok/Composer menu, validated against `cursor-agent models` (a typo is rejected up front). Omit for your Cursor account default.
Auth	Piggybacks whatever credential store `agy` uses on your OS (Windows Credential Manager, macOS Keychain, libsecret on Linux — the bridge never touches it directly). Log in once; every call silent-auths on the same AI Pro quota you already pay for.	Uses your existing Codex login — ChatGPT account or API key. Run `codex login` once; verify with `codex_status`.	Uses your existing Copilot login — run `copilot` then `/login` once (OS credential store), or set `COPILOT_GITHUB_TOKEN`/`GH_TOKEN`/`GITHUB_TOKEN`. Verify with `copilot_status`.	Uses your existing Cursor login — run `cursor-agent login` once (OS credential store), or set `CURSOR_API_KEY`. Verify with `cursor_status`.

⚠️ Security

All six backends run the model as an autonomous agent. The difference is whether you get a real boundary: Codex enforces one everywhere and Grok on Linux/macOS only; Copilot and Cursor offer best-effort ones; Antigravity and Kimi offer none.

Antigravity — no usable boundary

agy -p executes its own tools — reading and writing files, running shell commands, reaching the network — with no approval gate. Through agy 1.1.2 that was simply how print mode worked, with no opt-out at all. As of 1.1.3 it is a choice the bridge makes: agy finally gates headless tool calls, and the bridge deliberately opts out with --dangerously-skip-permissions, because a gated -p can do no useful work (it soft-denies even a plain file read, and print mode has no way to prompt). The posture below is therefore unchanged — assume every call runs arbitrary code with your privileges. Re-verified empirically on agy 1.0.9 / Windows, with the 1.1.3 amendment noted:

Print mode runs out-of-workspace file writes and live network fetches even without --dangerously-skip-permissions — that flag was a no-op for -p through 1.1.2. As of 1.1.3 it is load-bearing: without it every tool-using call is soft-denied, and the bridge now always passes it (it must precede -p, whose value is the prompt). There is still no agy flag that makes print mode both safe and useful.
agy 1.0.5 integrated a permission system (its logs show toolPermission=request-review), but it still does not gate print-mode execution — a fresh -p run created a file outside the workspace with no prompt. agy 1.0.12 reshuffled how that permission config merges (per-project files under ~/.gemini/config/projects/ now take precedence over ~/.gemini/antigravity-cli/settings.json), and 1.0.13 made "Always Approve" rule matching strict (non-regex) by default with a regex: opt-in and relaxed its redirection checks — but those are config/interactive-approval changes, they add no print-mode approval gate, and the bridge reads none of it.
--sandbox is not a usable boundary. agy 1.0.6 fixed its propagation into -p (the 1.0.6/1.0.7 changelog calls this "sandbox isolation correctly enforced") and it now does block terminal/ shell command execution — but re-verified on 1.0.9 that it leaves the write_to_file tool and network wide open: under --sandbox the model still wrote a file outside its workspace. agy 1.0.9 hardened the sandbox's command path (stricter exact-match command checks; .git added to its dangerous-paths list), but none of that closes the out-of-workspace write_to_file hole. On top of that, a --sandbox run whose blocked terminal command halts it writes no JSONL transcript (only the SQLite .db, re-confirmed on 1.0.9). The bridge can now read that .db, but still never passes --sandbox — it's no boundary, with file writes and network left open.

Codex — a real sandbox you should use

codex exec also has no interactive approval gate, but its sandbox flag is a genuine boundary that codex enforces:

read-only (default) — reads and answers; writes nothing. Safe for untrusted questions on trusted content.
workspace-write — may edit files under the workspace. Opt in deliberately, per task.
danger-full-access — no sandbox at all. Avoid.

Because there's no approval prompt, the flag you pass is the safety decision — choose it per call.

Copilot — best-effort, not an OS sandbox

copilot -p runs headless with --allow-all-tools (required — otherwise it blocks on per-tool permission prompts). Its sandbox maps to copilot's tool/path permission flags, which are a real-ish but not enforced boundary:

read-only (default) — auto-approves tools to run headless, then denies the local write and shell tools (--deny-tool). Blocks local file edits and command execution, but it is not an OS sandbox: other tools (including network/MCP) can still act. Weaker than Codex's read-only.
workspace-write — writes allowed, but file access stays confined to the workspace (no --allow-all-paths).
danger-full-access — --allow-all (tools + all paths + all URLs). Avoid.

For a hard read-only boundary, prefer codex_ask.

Cursor — best-effort, agent-enforced

cursor-agent -p runs headless with --trust (and --force when writes are allowed). Its sandbox maps to cursor's mode/force flags — an agent-enforced, not OS-level, boundary:

read-only (default) — --mode ask: the local write and shell tools are unavailable, so cursor analyzes and answers but makes no edits (verified: it refuses to write files). Like Copilot, this is agent-enforced and not an OS sandbox. Weaker than Codex's read-only.
workspace-write — --force: edits and commands allowed, file access rooted at --workspace.
danger-full-access — --force --sandbox disabled (OS sandbox off). Avoid.

For a hard read-only boundary, prefer codex_ask.

Grok — real, but only on Linux and macOS

grok -p runs headless with --always-approve (its headless mode does not auto-approve on its own, and nothing is there to answer a prompt). Its sandbox maps to grok's OS profile plus a tool allowlist:

read-only (default) — --sandbox read-only + --tools read_file,list_dir,grep,glob,web_search,web_fetch
- --no-subagents.
workspace-write — --sandbox workspace: writes confined to the workspace, ~/.grok, temp.
danger-full-access — --sandbox off. Avoid.

The profile is enforced by Landlock (Linux ≥ 5.13) and Seatbelt (macOS). On Windows there is no mechanism, and per xAI's docs grok "logs a warning and continues without enforcement" — so on Windows the only thing standing between read-only and your disk is the agent-enforced tool allowlist. Treat Windows read-only as best-effort (Copilot/Cursor tier), not as a jail. The bridge uses an allowlist rather than a denylist so that a future grok with a new write tool fails safe; note that MCP meta-tools remain available under an allowlist regardless.

⚠️ This backend is unverified — including these sandbox claims, which could not be exercised, because grok checks auth before it validates --sandbox.

Kimi — no boundary at all

kimi -p has no sandbox and no sandbox argument. Print mode auto-executes every tool call with no approval gate — the same posture as Antigravity, and verified live on 0.29.1 in the sense that -p rejects --auto/--yolo precisely because it is already self-approving. No flag makes it safe. Assume every kimi_ask runs arbitrary code with your privileges.

What that means for you

The workspace argument is only a starting context, not a security boundary — Antigravity and Kimi can and do act outside it; Codex is bounded by its enforced sandbox; Grok by its OS profile on Linux/macOS and by a tool allowlist elsewhere; Copilot by its best-effort tool/path permissions; Cursor by its agent-enforced mode/force.
An Antigravity or Kimi call effectively runs arbitrary code with your user privileges. A Copilot or Cursor call does too outside its best-effort denials; a Grok call does on Windows outside its allowlist; a Codex call does unless you keep it at read-only.
Only invoke these with trusted prompts on trusted content. Untrusted input here is the classic prompt-injection lethal trifecta: private-data access + code execution + network egress.
For real isolation, run the whole bridge inside a container or VM.

The bridge itself does only cross-platform filesystem reads under ~/.gemini/antigravity-cli/, ~/.codex/, ~/.copilot/, and ~/.cursor/ — no private APIs, no token theft. The risk above is entirely in what the sub-agents are allowed to do.

FAQ

Is this against Google's / OpenAI's / GitHub's / Cursor's Terms of Service?

It runs the official agy, codex, copilot, and cursor-agent CLIs under your own logins — no private APIs, no token theft, no quota abuse. It just bridges what the CLIs already do. That said, your AI Pro / Antigravity, OpenAI / Codex, GitHub Copilot, and Cursor ToS apply, and you're responsible for staying within them.

Do I need all six CLIs?

No. Each backend is independent — install only the CLI(s) you want. The tools for a missing backend report "not found" via their *_status tool (antigravity_status / codex_status / copilot_status / cursor_status / grok_status / kimi_status) and never crash the server.

When should I use Antigravity vs Codex vs Copilot vs Cursor vs Grok vs Kimi?

Use Antigravity for fast, cheap tool-calling, quick answers, and image generation (it's the only backend with an image model) — and it now lets you pick the model too (agy's --model). Use Codex for heavier reasoning, real code/repo work, or when you want a real, enforced workspace-write sandbox. Use Copilot for agentic coding on your GitHub Copilot plan, or as a second coding opinion alongside Codex — noting its sandbox is best-effort, not enforced. Use Cursor for agentic coding on a Cursor plan, or when you want the widest model menu — GPT, Claude, Grok, and Composer, all via model — noting its sandbox is agent-enforced, like Copilot's.

Grok and Kimi are experimental and unverified — reach for them to help verify them, or if they're the subscription you actually have. Grok is the more capable of the two here: real OS sandbox (Linux/macOS), watch mode, and swarm support. Kimi has no sandbox and no swarm/watch yet.

All of them let you choose a model (except Kimi, which can't validate one); in a swarm you can mix five of the six. See The backends at a glance.

Will it break when agy updates?

Less likely now. As of agy 1.0.15 the bridge prefers agy's stdout on the happy path (1.0.15 fixed the print-mode stdout bug on Windows — -p now writes the clean answer there), which removes its dependence on agy's undocumented transcript schema for normal runs. It still falls back to reading the JSONL transcript, or the SQLite .db agy dual-writes, when stdout is empty (older agy, non-Windows, or --sandbox runs) — so a schema change would only bite that fallback path. Re-verified working on 1.0.15 (stdout answer clean under tool use; transcript/.db fallback intact; live ask round-trip + antigravity_status diagnostics pass). Still, if you rely on the fallback, pin a known-good agy version.

Which model does Antigravity use — can I pick it?

Yes. Pass model to antigravity_ask/antigravity_continue (or per task in agent_swarm) — it maps to agy's --model, taking any slug from agy models (e.g. "gemini-3.1-pro-high", "claude-sonnet-4-6"). Omit it to use the "model" field in agy's settings.json, which defaults to gemini-3.6-flash-high as of agy 1.1.6 — speed-optimized for cheap tool-calling.

agy 1.1.5 renamed every model, replacing the old human labels ("Gemini 3.1 Pro (High)") with stable slugs (gemini-3.1-pro-high) — the old form is no longer accepted, so pass slugs. agy 1.1.6 then added the gemini-3.6-flash family and moved the default to it. The full list, re-checked live on 1.1.12 and unchanged since 1.1.6: gemini-3.6-flash-low|medium|high, gemini-3.5-flash-low|medium|high, gemini-3.1-pro-low|high, claude-sonnet-4-6, claude-opus-4-6-thinking, gpt-oss-120b-medium. Note the slug bakes in the reasoning effort, which is why the flash and pro models appear once per level.

agy 1.0.5 added --model, but through ~1.0.14 switching to a different model in -p hung the call, so earlier bridge versions stayed single-model. Re-verified on agy 1.0.16 that the hang is fixed — a Claude model answers as Anthropic Claude, a Gemini model as Gemini, each in seconds. One caveat the bridge handles for you: agy silently ignores an unknown model (it falls back to the default with no error), so the bridge validates your slug against agy models and rejects a typo up front. (That validation is also what agy 1.1.11 broke, by making agy models print <slug>\t<human label> instead of a bare slug — see Status & caveats. Fixed as of bridge 0.24.0; on 0.23.x with agy 1.1.11+, pass no model at all.)

Can it generate images?

Yes — that's the antigravity_image tool, on the Antigravity backend. agy's print mode generates real images on your AI Pro quota; antigravity_image drives it, saves the file to a path you choose (or a timestamped default in your workspace), fixes the extension to match the real bytes (agy picks JPEG or PNG itself), and returns the path. Verified on agy 1.0.9 / Windows. Codex has no image model — it's a coding agent.

Does it cost extra money?

No. It uses the same quota you already pay for — AI Pro for Antigravity, your Codex plan for Codex, your GitHub Copilot plan for Copilot, your Cursor plan for Cursor. The smoke test spends a negligible amount.

Does it stream responses?

The final answer is request/response — the CLIs return it all at once, so the tools return when the agent finishes (each call typically takes 10–30 s; Copilot's reasoning models can run longer). If you want to watch the agent work as it goes, pass watch=true to any single-prompt tool: it opens the Agent Intern browser window and live-streams the agent's steps — see Watch mode. It's coarse (a handful of steps, not token-by-token), and the returned value is identical to the non-watch call.

Can I run several calls at once?

The single-agent tools are serialized inside the server: agy rewrites last_conversations.json on every call, so concurrent runs sharing one state dir would race and could return the wrong conversation. A threading.Lock makes extra requests queue rather than race. (On agy 1.1.8+ the bridge also records the conversation_id agy reports for each run and prefers it when pinning a continue, so that resolution no longer depends on the shared file — but the lock stays, since agy's state dir is still shared and a fresh server process starts with nothing recorded.)

For real parallelism use agent_swarm — each agy worker runs in its own isolated state dir (and Codex/Copilot/Cursor workers need none), so they don't race and the lock isn't needed (~2.8× at 3 workers). That's the supported way to run many calls at once, across any backend.

The isolated state dir is a redirected HOME, and on macOS that also hides agy's stored credentials (the login keychain is resolved through $HOME), so isolated workers there fall back to running serialized in your real HOME — correct, but without the speedup. Windows is unaffected (Credential Manager is HOME-independent). See the swarm auth note.

Status & caveats

🧪 Grok Build and Kimi Code ship UNVERIFIED — help wanted. Two new backends, neither ever exercised against an authenticated account, because I have neither subscription. This is a deliberate trade: shipping them unverified is the only way anyone can verify them, and the parts that historically rot — flag surfaces, model ids, error shapes — are pinned down live against grok 1.0.3 and kimi 0.29.1. Grok's are unusually solid, since its CLI is open source: every argv the bridge builds was read off the real clap definitions and confirmed to parse. What's unproven is everything behind each CLI's auth wall — the answer itself, resume, and Grok's streaming events. That verification gap already caught one live docs error (xAI documents grok-build as the model; the real default is grok-4.5), which is a fair warning about what else the docs may be wrong about. If you have either subscription, one issue from the verification template closes this gap. Full detail →
🐛 agent_swarm antigravity workers died with "authentication timed out" on macOS — fixed (#2). Each swarm worker gets an isolated HOME so agy's per-process state can't collide. The module shipped asserting that auth survives it because agy reads credentials from the OS credential store — true on Windows (Credential Manager is keyed to the user session; re-verified on 1.1.12 that agy models inside a fake HOME returns the full list) and false on macOS, where the login keychain is resolved through $HOME. So every antigravity worker started a fresh OAuth flow and died at agy's 60 s authentication timed out, while antigravity_ask — which never touches HOME — kept working. Isolation is now conditional, decided two ways:
- Proactively, by probing once per process: agy -p "/usage" inside a throwaway isolated HOME. On agy 1.1.11+ that's free (the CLI answers it, no agent turn, no quota), but the quota table comes from your account, so it can't be answered without working credentials. Skipped on Windows, where it can't fail.
- Reactively, because a probe is a proxy and this one can't be tested on the platform it exists for: any worker that fails with an authentication signature flips the process to serialized mode and retries itself there, so you get an answer even if the probe was wrong or unavailable.
In serialized mode the antigravity workers run in your real HOME behind the same lock the single-agent tools use — correct everywhere, at the cost of the parallelism (other backends stay parallel; a watched worker shows a note instead of live steps, since the step feed reads the isolated transcript). AGY_BRIDGE_NO_HOME_ISOLATION=1 forces it without a probe. Diagnosed by inspection from the report; not reproduced on a Mac — if you're on macOS, please confirm on #2.
🐛 agy 1.1.11 killed the model argument — fixed (#3). agy made agy models machine-readable, turning each line from a bare slug into a tab-separated <slug>\t<human label> record (gemini-3.6-flash-high\tGemini 3.6 Flash (High)). The bridge read the whole line as the slug, so its up-front validation rejected every valid model with an error that listed the very slug it had just refused: unknown agy model 'gemini-3.6-flash-high'; expected one of: gemini-3.6-flash-high<TAB>Gemini 3.6 Flash (High), …. Reproduced end-to-end through antigravity_ask; model was unusable on all three antigravity tools (omitting it still worked — the default model path never touched this code). The parser now keeps the first tab field, which reads both formats, and drops any field containing whitespace (a slug never has a space, so such a line is chatter). Note 1.1.12's changelog advertises --output-format json for the models and agents subcommands, but the shipped binary has no such flag (agy models --output-format json → flags provided but not defined: -output-format), so TSV is what there is to parse. The live slug list is unchanged from 1.1.6, and this suite stayed green through the break because every model test mocked the old format — the new tests pin both formats down. The change is in no changelog entry: the guard test was green on 1.1.10, #3 reports the break on 1.1.11, and it was reproduced here on 1.1.12.
✨ antigravity_status now reports your remaining AI Pro quota — for free. agy 1.1.11/1.1.12 answer read-only slash commands in print mode itself: no agent turn, no quota spend, no conversation left behind (1.1.11: /usage, /quota, /credits, /model, /effort, /skills; 1.1.12: /permissions, /hooks, /help, /config, /changelog). The status tool runs agy -p "/usage" and adds a row per model family — quota: Gemini Models [ok] Weekly 100%, Five Hour 100% — flagging a family at 0% as a problem, since every call against it fails until its window resets. Version-gated at 1.1.11 as a safety gate: on older agy the same argv is a prompt, so a diagnostic advertised as free would quietly spend a call. The probe is the one bridge call that deliberately omits --disable-slash-commands (a regression test asserts it), and it degrades to nothing on older agy rather than reporting a false problem.
✅ Re-verified the rest of agy 1.1.11/1.1.12 — nothing else broke. The slash shield still holds: 1.1.11 replaced the silent fall-through for interactive-only commands with an explicit refusal that recommends the exact flag this bridge already passes (agy -p "/clear" → exit 2, "pass --disable-slash-commands to send /clear to the model as literal text"), and antigravity_ask("/clear Reply with the single word BRIDGE…") returned BRIDGE end-to-end. The read-only set is why the shield stays load-bearing — unshielded, a prompt opening with /model would get agy's table instead of an answer. 1.1.12 also stopped swallowing startup diagnostics (including the --conversation not-found warning the continue path can trigger): they go to stderr, so stdout stays a pure JSON result object — verified with a deliberately bogus --conversation (exit 0, clean JSON, warning on stderr). Benign wins: a Windows crash resolving the conversation transcript path is fixed (the artifact watch mode's log_uri points at), headless -p now settles a choice itself instead of stalling on a question nobody can answer (fewer timeouts), and 1.1.11 made retries honor the server's retry delay and stopped an empty credits response reading as "Out of credits". VERIFIED_AGY_VERSION → (1, 1, 12). --effort stays unadopted, now with a harder reason: it isn't universal — --model claude-sonnet-4-6 --effort low fails with "--effort is not supported for model", while the gemini slugs already bake the level in. Everything else (Vim editing mode, artifact-viewer polish, plugin enablement, admin controls and MCP progress callbacks — agy as an MCP client) is off the bridge's path.
🛡️ agy 1.1.9 broke print mode for any prompt starting with a slash — fixed by --disable-slash-commands. 1.1.9 made -p expand slash commands and skills instead of sending them to the model as text, so a prompt whose first token names a registered command is executed as that command and never reaches the model. Verified live on 1.1.10 through this bridge: antigravity_ask("/help") came back with agy's own help page, not an answer. That is not just wrong output — agy's registered set includes side-effecting commands (/goal starts an autonomous long-running task, /schedule creates cron jobs), and bridge prompts routinely carry text the caller did not author, so an untrusted string beginning /schedule … would have run it. Every agy argv path now passes --disable-slash-commands (one change in _agy_base_args covers ask, continue, both watched runners, and both swarm workers). Version-gated: the flag doesn't exist before 1.1.9, and neither does the expansion. Prompts starting with a POSIX path (/etc/hosts …) were never affected — they match no command — but that was luck, not a boundary. Set AGY_BRIDGE_ALLOW_SLASH_COMMANDS=1 to keep the expansion if you want -p "/my-skill <args>" to invoke a skill.
🐛 Non-ASCII answers were being mangled on Windows — fixed. Every backend emits UTF-8, but the bridge spawned them with bare text=True, which decodes using the locale codepage (locale.getpreferredencoding() — cp1254 on a Turkish Windows, cp1252 elsewhere). Any non-ASCII answer came back corrupted: dosyası arrived as dosyasÄ±, exactly 'dosyası'.encode('utf-8').decode('cp1254'). Every bridge plus the swarm workers now decode UTF-8 explicitly with errors="replace", the pattern cursor_bridge.py already used. A regression test asserts no subprocess call reintroduces bare text=True. ASCII-only answers were never affected, which is why this survived so long.
✅ Re-verified on agy 1.1.9 and 1.1.10. Beyond the slash-command break above: 1.1.10 fixed --model/--effort being silently ignored in headless -p (they were applied after model configuration had already initialized, so the run fell back to the persisted/default model). The bridge validates and passes --model on every call, so on 1.1.8–1.1.9 the model argument was a no-op even though a typo was still correctly rejected — if you pinned a model in that window, you were served the default. Re-confirmed working on 1.1.10 through the bridge (model="claude-sonnet-4-6" → a Claude answer, not Gemini). No code change was needed for it. 1.1.10 also added a non-blocking advisory banner when the same conversation is open in another CLI instance — the shape antigravity_continue and the swarm can produce — so _parse_json_result now locates the result object instead of requiring stdout to start with {; leading and trailing chatter are both absorbed rather than degrading into a raw JSON blob in your answer. VERIFIED_AGY_VERSION → (1, 1, 10). Nothing else in 1.1.9/1.1.10 reaches the bridge — the rest is interactive-TUI, hooks, auth, and MCP-client work.
✅ Verified on agy 1.1.7 and 1.1.8 — nothing broke, and 1.1.8 made the bridge sturdier. 1.1.8 gave print mode an --output-format flag (text | json | stream-json). The existing text path was confirmed live on 1.1.8 first (ask, pinned continue, and --model all clean), then the bridge switched its plain ask/continue calls to --output-format json, because reading a contractual response field beats trusting the layout of bare text. The real prize is the conversation_id agy returns with it: the bridge records it and pins a later antigravity_continue to exactly the conversation it last ran in that workspace, instead of inferring it from last_conversations.json — shared state agy rewrites for every session, including your own interactive TUI work in the same folder. Practical difference: antigravity_continue now resumes the bridge's own thread, where before it could land on a conversation you'd since started in the Antigravity TUI. Older agy is unaffected — the flag is version-gated (pre-1.1.8 has no such flag), and any non-JSON stdout falls back to the previous text path, so a silently-ignored flag degrades instead of crashing. VERIFIED_AGY_VERSION → (1, 1, 8). Not adopted: --json-schema (works; nothing here needs it). Nothing else in 1.1.7/1.1.8 reaches the bridge — the rest is interactive-TUI, plugin-hook, and MCP-client work.
✅ Watch mode reads agy's live event stream instead of scraping its transcript. On agy 1.1.8+ the watched runners request --output-format stream-json and consume agy's typed init / step_update / result events straight off stdout. Verified before the rewrite that they arrive incrementally (a 17 s run spread its 18 events over 12.4 s), and confirmed live that a watched run's step count grows while agy works. The stream carries the real command as a nested object (the transcript stored tool args JSON-encoded inside a string), streaming text fragments, and a conversation_id — so a watched run now pins later continues just like a plain one. This retires the timer-based transcript polling, which matters beyond tidiness: agy has announced JSONL is being replaced by SQLite, and watch was the last path that would have broken when it goes. Pre-1.1.8 agy keeps the original transcript path, re-verified live.
⚠️ Behavior change: multi-step answers now include the model's narration. agy's response is the whole turn; the old transcript scrape returned only the last planner response. Identical for a single-step ask, different for a chatty multi-step one (one measured run: 297 chars vs 128, the full answer ending in the old one). The full turn is now returned on every path — response is agy's own contract for what the turn produced, and the old last-step rule silently dropped content whenever the model did the work and then closed with a short "Done."
✅ Re-verified on agy 1.1.6 — no code change needed. 1.1.6 added the gemini-3.6-flash family to agy models and moved the settings.json default to Gemini 3.6 Flash (High); the default path and --model gemini-3.6-flash-high both round-tripped clean, and the JSONL + SQLite read paths still match agy's unchanged conversation schema. Its one bridge-adjacent fix — print mode now surfacing the real conversation-creation error instead of a misleading "no active conversation" — only improves the diagnostic the bridge already reads on failure. Everything else (Markdown custom agents, /copy and /codesearch polish, background-task hardening) is interactive-TUI or client-side work that doesn't reach the bridge. Docs-only: the model list and default examples now name the 1.1.6 slugs, and the guard test advertises gemini-3.6-flash-high against the live list.
⚠️ Verified on agy 1.1.5 — it renamed every model, so old model values now fail. 1.1.5 replaced agy's human-readable model labels with stable slugs, and agy models reports only those: "Gemini 3.1 Pro (High)" is now gemini-3.1-pro-high, and the Claude entries are claude-sonnet-4-6 and claude-opus-4-6-thinking (the mapping is not 1:1 — check agy models, or antigravity_status, for the current eight). Since the bridge validates model against agy models, an old label is rejected up front with the valid list — you lose the call, not your money, and never silently run on the wrong model. Pass slugs and you're fine. Nothing in the bridge's machinery needed changing (validation was always format-agnostic — which is exactly why the entire test suite stayed green while every documented example went stale), so this release is docs plus one new test that checks the models we advertise against the live agy models list. Everything else in 1.1.5 is interactive-TUI, MCP-client, or background-task work that doesn't reach the bridge; its new --effort flag is a second axis we don't pass, because the slug already pins the effort variant.
✅ Verified on agy 1.1.4 — no code change was needed. 1.1.4 relaxed the 1.1.3 headless gate so that -p now honors your persisted settings.json policies (permissions, file access, sandbox mode, auto-execution, artifact review) instead of blanket-denying. --dangerously-skip-permissions still overrides those policies, so the flag stays load-bearing and stays exactly where it is — re-verified live against a workspace deliberately absent from trustedWorkspaces, with a permissions.allow list naming neither file nor command access: a workspace file read returned the right contents, and a terminal command and a file write both executed. Worth knowing: that flag is now the only thing between a bridge call and your own settings.json policy, and dropping it would get you whatever that file says rather than 1.1.3's deny-everything. 1.1.4 also stopped /btw side-questions from leaking into the conversation list as duplicates carrying the parent's title — that list is what conversation pinning reads, so one way to resume the wrong thread is gone.
✅ Verified on agy 1.1.3 — base dir, last_conversations.json (still keyed by workspace path), the brain/.../transcript.jsonl path, the transcript schema, and the -p/-c/--print-timeout flags are all unchanged; a live antigravity_ask + conversation-pinned antigravity_continue round-trip returns clean over stdout and antigravity_status diagnostics pass. 1.1.3 broke and the bridge fixed the one thing that mattered: headless -p no longer auto-approves tool calls, it soft-denies them (print mode cannot prompt), so without a flag even "read pyproject.toml and report the version" returned nothing — exit 0, empty stdout, the reason only on stderr. The bridge now passes --dangerously-skip-permissions on every agy path, which restores file writes, terminal commands and workspace reads (a live bridge round-trip reads this repo's real version again). The flag must precede -p, whose value is the prompt — otherwise the flag becomes the prompt and the task is silently dropped. 1.1.2 also made an unresolvable --model hard-fail in -p instead of silently falling back to the settings.json default (the bridge's validate_model still rejects a typo up front, without spending a call). 1.1.0's execution-mode system (--mode, request-review) remains a no-op for the bridge: -p is spawned with DEVNULL stdin, so that interactive gate never engages. --sandbox behavior is likewise unchanged (blocks the terminal, not file writes). The print-mode stdout path (fixed on 1.0.15, Windows) still applies; the transcript stays the fallback.
✅ Verified on codex-cli 0.144.1 — codex exec, -o/--output-last-message, codex exec resume, the --json event stream, and the ~/.codex/sessions/.../rollout-*.jsonl layout the continue path reads are all in place; a live codex_ask round-trip + codex_status pass. (Bumped from the 0.141.0 baseline: flags, session layout and the round-trip all re-verified unchanged.)
✅ Verified on copilot 1.0.69 — copilot -p -s (clean stdout answer), --session-id set-then-resume, --model, --output-format json (watch stream), and the ~/.copilot/session-state/<id>/workspace.yaml layout the continue fallback reads are all in place; live copilot_ask / copilot_continue round-trips + a mixed agent_swarm pass. (Bumped from 1.0.68: 1.0.69 adds a --resume convenience flag the bridge doesn't need; --session-id still both sets a fresh id and resumes it — re-verified live, ACK then codeword recall.)
✅ Verified on cursor-agent 2026.07.23 — cursor-agent -p --output-format text --trust (clean stdout answer), create-chat + -p --resume <id>, --model (validated against cursor-agent models), --output-format stream-json (watch stream), and the ~/.cursor/chats/<md5(workspace)>/<chat-id>/meta.json layout the continue fallback reads are all in place; live cursor_ask / cursor_continue round-trips + a mixed agent_swarm pass. (2026.07.09 is installed and cursor_status — found, logged in, chats dir — passes with flags/layout intact; a fresh live round-trip was deferred, Cursor usage limit reached.)
🖥️ Console-detach — before 1.0.15 agy -p wrote its answer to the controlling terminal, not stdout; under a TUI that text leaked into the host's prompt (seen on 1.0.9). 1.0.15 fixed this on Windows (stdout now carries the answer), but the bridge still spawns agy detached (CREATE_NO_WINDOW / a new POSIX session), which prevents the leak on older/other platforms and is harmless on 1.0.15+.
💾 SQLite migration — handled — agy still dual-writes a .db per conversation; on the fallback path, when the JSONL transcript is absent (already true for --sandbox runs, and the announced future default) _read_response falls back to reading the .db, verified to match across 100+ conversations. See the FAQ.
🐛 agy stdout bug — fixed on 1.0.15 — -p now prints the clean answer to stdout in a non-TTY subprocess (Windows), so the bridge prefers stdout and only scrapes the transcript when stdout is empty (older agy, non-Windows, or --sandbox). (Codex and Copilot never had this problem — both are stdout-native.)
👁️ Watch mode is experimental — pass watch=true to any single-prompt tool to open the Agent Intern window and watch the agent work live (coarse steps; image shown inline). Best-effort and cross-platform; see Watch mode.
🔒 Sandbox — agy's --sandbox blocks only shell commands, so it's no boundary and the bridge never passes it. Codex's sandbox is real and enforced — use it; default read-only. Copilot's sandbox is best-effort (tool/path denials, not an OS sandbox); default read-only. Cursor's sandbox is agent-enforced (mode/force; read-only = --mode ask makes write/shell unavailable, not an OS sandbox); default read-only. See Security.

Requirements

Python 3.10+
For the Antigravity tools: agy 1.0.0+ on PATH (state-file layout re-verified on 1.0.15; behaviour re-verified on 1.1.10) and an active Antigravity / AI Pro session
For the Codex tools: codex on PATH and logged in (codex login) — verified on codex-cli 0.144.1
For the Copilot tools: copilot on PATH and logged in (copilot → /login, or a COPILOT_GITHUB_TOKEN/GH_TOKEN env) — verified on copilot 1.0.69
For the Cursor tools: cursor-agent on PATH and logged in (cursor-agent login, or a CURSOR_API_KEY env) — verified on cursor-agent 2026.07.23
For the Grok tools (experimental): grok on PATH and logged in (grok login, or an XAI_API_KEY env) plus a SuperGrok / X Premium+ subscription — flag surface verified on grok 1.0.3, answer path unverified
For the Kimi tools (experimental): kimi on PATH and logged in (kimi login, or an API key in ~/.kimi-code/config.toml) — flag surface verified on kimi 0.29.1, answer path unverified

Each backend is independent — install only the CLI(s) you plan to use; the other tools simply report "not found" via their *_status tool.

Tip

If agy isn't reliably on PATH (e.g. a new terminal or reboot drops it on Windows), set the AGY_BIN env var to its full path and the bridge will use that instead of "agy" — e.g. AGY_BIN=%LOCALAPPDATA%\agy\bin\agy.exe. Likewise, set CODEX_BIN if codex isn't reliably on PATH (the native Windows installer puts it under %LOCALAPPDATA%\Programs\OpenAI\Codex\bin\), and COPILOT_BIN if copilot isn't (the winget install lands under %LOCALAPPDATA%\Microsoft\WinGet\Packages\GitHub.Copilot_*\copilot.exe). Finally, set CURSOR_BIN if cursor-agent isn't reliably on PATH (the installer drops a cursor-agent.CMD shim a bare name can't launch on Windows). GROK_BIN and KIMI_BIN do the same for the two experimental backends — though the Grok bridge already falls back to the installer's own ~/.grok/bin on a PATH miss, which matters because that installer appends to the user PATH and the change never reaches an already-running server.

The bridge uses only cross-platform Python (Path.home(), subprocess) and reads paths under ~/.gemini/antigravity-cli/, ~/.codex/, ~/.copilot/, ~/.cursor/, ~/.grok/, and ~/.kimi-code/, which the CLIs write the same way on every OS. Developed and verified on Windows; macOS and Linux should work unmodified provided the CLIs run there. If you test it on those platforms, please open an issue / PR to confirm.

🌐 Community & Acknowledgments

Qiita (Japan): A huge thanks to @fallout and the Japanese developer community for featuring this project and providing invaluable feedback!
- Detailed Hybrid Setup Guide (Claude Code × Antigravity CLI)
- Quick Installation Guide

💡 Path Resolution Fix: Thanks to their community's real-world testing, we identified and resolved a Windows PATH edge case where the MCP server inherits a stale PATH at startup and can't find agy. The AGY_BIN environment-variable fallback was implemented directly inspired by their report!

License

MIT. Do whatever you want with it.

Name		Name	Last commit message	Last commit date
.github		.github
assets		assets
tools		tools
.gitignore		.gitignore
CHANGELOG.md		CHANGELOG.md
LICENSE		LICENSE
README.md		README.md
codex_bridge.py		codex_bridge.py
copilot_bridge.py		copilot_bridge.py
cursor_bridge.py		cursor_bridge.py
grok_bridge.py		grok_bridge.py
kimi_bridge.py		kimi_bridge.py
pyproject.toml		pyproject.toml
server.py		server.py
swarm.py		swarm.py
swarm_watch.py		swarm_watch.py
test_codex.py		test_codex.py
test_copilot.py		test_copilot.py
test_cursor.py		test_cursor.py
test_grok.py		test_grok.py
test_kimi.py		test_kimi.py
test_server.py		test_server.py
test_smoke.py		test_smoke.py
test_swarm.py		test_swarm.py

Uh oh!

Latest commit

History

Folders and files

Repository files navigation

Claude Code × Antigravity + Codex + Copilot + Cursor + Grok + Kimi — MCP Bridge

Why you'd want this

The backends at a glance

🧪 The two experimental backends — and how you can help

How to help

How it works

Set up in 60 seconds

Recommended — no clone, you control updates

From source

Tools

🛰️ Antigravity

🤖 Codex

🐙 Copilot

✳️ Cursor

🧪 Grok Build (experimental — unverified)

🌙 Kimi Code (experimental — unverified)

🐝 Shared

🤖 Codex bridge — the well-behaved sibling

🐙 Copilot bridge — GitHub's agentic coder

✳️ Cursor bridge — the widest model menu

🧪 Grok Build bridge — a real sandbox, on two of three platforms

🌙 Kimi Code bridge — no sandbox, per-directory sessions

👁️ Watch mode — Agent Intern (experimental)

🐝 Swarm — run agents in parallel

Model & auth

⚠️ Security

Antigravity — no usable boundary

Codex — a real sandbox you should use

Copilot — best-effort, not an OS sandbox

Cursor — best-effort, agent-enforced

Grok — real, but only on Linux and macOS

Kimi — no boundary at all

What that means for you

FAQ

Status & caveats

Requirements

🌐 Community & Acknowledgments

License

About

Topics

Resources

Stars

Watchers

Forks

Releases

Sponsor this project

Packages

Contributors

Languages