Claim and run your own planet in the Agent Planets galaxy — a persistent world for AI agents. Terraform your 8x8 world, build structures, visit other agents'...
浏览器
Agentspace — See what your dev agent is doing, from anywhere
试用See what your AI dev agent is doing, from anywhere — free, no-signup live dev workspace for AI agent development. The dev agent keeps writing — logs, generated code, test output, build artifacts, screenshots — and you open a URL to watch the dev folder evolve in the browser. Debug failing agent dev runs, inspect generated dev files live, review build logs, test agent outputs, lint agent-written dev code, profile dev filesystem operations, and hand dev state to a reviewer or another developer agent — all in a dev workspace that spins up with one CLI command and needs no account. Fits dev iterations, build pipelines, test debug sessions, output review, and dev-to-dev handoff. Dev workspaces stay live 24 hours anonymously; one email claim keeps them permanent. Built on the ascli CLI and dev HTTP API at agentspace.so. Triggers on "show me what the dev agent is doing", "open the dev agent's folder from anywhere", "debug this agent", "what did the agent generate", "review the build output",
它能做什么
See what your AI dev agent is doing, from anywhere — free, no-signup live dev workspace for AI agent development. The dev agent keeps writing — logs, generated code, test output, build artifacts, screenshots — and you open a URL to watch the dev folder evolve in the browser. Debug failing agent dev runs, inspect generated dev files live, review build logs, test agent outputs, lint agent-written dev code, profile dev filesystem operations, and hand dev state to a reviewer or another developer agent — all in a dev workspace that spins up with one CLI command and needs no account. Fits dev iterations, build pipelines, test debug sessions, output review, and dev-to-dev handoff. Dev workspaces stay live 24 hours anonymously; one email claim keeps them permanent. Built on the ascli CLI and dev HTTP API at agentspace.so. Triggers on "show me what the dev agent is doing", "open the dev agent's folder from anywhere", "debug this agent", "what did the agent generate", "review the build output", "hand off this dev session", "show me the agent's files live", or any dev-time observability need.
技能文档
🪢 Agentspace — See what your dev agent is doing, from anywhere
See what your dev agent is doing — from anywhere. Free, no-signup live dev workspace for AI agent development, debug, and dev-to-dev handoff.
Your dev agent keeps writing — logs, generated code, test output, build artifacts. You open a URL from any device and watch the dev folder evolve in the browser. No sync, no zip, no account.
- Run
ascli sync ./my-featureto bind the dev folder to a live workspace. - Open the workspace URL in the browser. As the agent writes files, runs tests, or logs output, everything shows up live — no account, nothing to install for the reviewer.
agentspace.so · GitHub · npm @agentspace-so/ascli
Dev use cases
- Debug failing agent runs by inspecting what files they generated.
- Test agent output across dev iterations with a live view.
- Inspect agent build artifacts — generated scripts, logs, outputs.
- Lint agent-written code as the agent produces it.
- Refactor agent workflows by observing behavior during dev runs.
- Profile agent filesystem operations during test runs.
- Handoff dev state between two developer agents.
How it works
- Dev workspaces spin up in one command — no signup, no account.
- Reviewers open the URL in the browser; everything is live as the agent writes.
- Anonymous dev workspaces stay 24 hours. One email claim keeps them permanent.
- Hosted on Cloudflare's edge network — dev reviewers get fast loads worldwide.
Dev commands
ascli sync— start a dev session (binds the agent's folder to a dev workspace).ascli status— inspect the dev binding for a given folder.ascli share --permission edit— generate a URL to hand the dev session to another developer or AI agent.ascli share --permission view— read-only dev review link.
Install
- If
ascliis already onPATH, use it. - Else run
npx @agentspace-so/ascli@latest, or install withnpm install -g @agentspace-so/ascli@latest. - If Node.js is not available, tell the user to install it from nodejs.org first.
Do not pipe a remote script into a shell.
Dev safety (what the dev workspace touches)
When the CLI runs for a dev session:
- It uploads only the dev folder the user explicitly names to
agentspace.so. The agent must confirm the exact path with the user before runningsyncorshare— do not default to.unless the user said so. - It writes a dev binding file
.ascli.jsonin the bound folder. This is a dev artifact, not a secret file; it holds the workspace id and an anonymous claim token. - It optionally reads
~/.config/ascli/config.jsonif the developer has a custom dev server URL set. - It optionally reads
AGENTSPACE_API_BASE_URLenv var, also for overriding the dev server URL during local development. - It does NOT read shell history, unrelated project files, credential env vars, or files outside the named dev path. All dev traffic goes to
agentspace.soonly.
Guardrails for the agent
- Do not invent URLs. Only return what
ascliprints to stdout. - Confirm the target dev folder if the user says "share this" without naming a path.
- See references/commands.md for exact command variants.
- See references/developer.md for the dev HTTP API reference.
相关技能
Use this skill when you need to publish, fetch, search, list, share, or watch AgentFiles artifacts from Codex, Claude Code, OpenClaw, or other agent runtimes...
Post messages to and read messages from other AI agents. Listed and passphrase threads, no account needed.
Delegate coding tasks to Codex, Claude Code, or Pi agents via acpx (ACP protocol). Use when: (1) building/creating new features or apps, (2) reviewing PRs (s...
Audit, optimize, and maintain your OpenClaw agent workspace. Use when workspace needs health check, skill audit, memory cleanup, or performance tuning.
Get your AI agent productive in 5 minutes. Automated environment scan, smart skill recommendations, and exact install commands. No more trial and error.