Solid Agent — Testing skill for Claude Code
Open spec + reference implementation for composing coding agents (Claude, Codex, AmpCode, Cursor, Pi, …) with runtimes (Local, E2B, Daytona, Modal, …).
How to install Solid Agent
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open ya-luotao/solid_agent and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Solid Agent does
Open spec + reference implementation for composing coding agents (Claude, Codex, AmpCode, Cursor, Pi, …) with runtimes (Local, E2B, Daytona, Modal, …). ACP-aligned, draft.
Alternatives in Testing
- Spec Kitty — Spec-Driven Development for serious software developers 1.6k ★
- Spec Driven Develop — Spec-driven development workflow for AI coding agents: architecture-first planning, task decomposition, GitHub 975 ★
- Spectra App — Spec-driven development for coding agents — a desktop app, a CLI, and skills for Codex, Claude Code, Cursor, C 735 ★
README
Solid Agent
**Status:** Early draft — open for discussion. Pre-alpha; no published gem yet.
**Solid Agent** is an open protocol — built on top of the [Agent Client Protocol (ACP)](https://agentclientprotocol.com/) — that wraps coding-agent SDKs and sandbox runtimes so they compose. A Ruby reference implementation ships alongside the spec.
SolidAgent.run('add tests for the auth module', agent: :claude, sandbox: :e2b)
The protocol specifies how to wrap an agent SDK so it speaks ACP, how to wrap a sandbox SDK so it fulfils ACP's client-side methods (`fs/*`, `terminal/*`), and adds a small set of runtime lifecycle methods (`runtime/snapshot`, `runtime/pause`, `runtime/resume`, `runtime/manifest_apply`) that ACP does not yet specify.
┌────── Agent ──────┐ ┌────── Runtime ──────┐
│ Claude Code │ │ Local │
│ Codex │ │ E2B │
│ AmpCode │ ─ ACP ─ │ Daytona │
│ Cursor │ │ Modal │
│ Pi │ │ Cloudflare │
│ ACPDirect │ │ Vercel │
│ … any ACP agent │ │ Runloop / Blaxel │
└───────────────────┘ │ OpenAI built-in │
│ … any ACP runtime │
└──────────────────────┘
Why this exists
Coding-agent SDKs and sandbox runtimes have proliferated. Composing them by hand for every project means re-writing the same glue — translating between each SDK's native event vocabulary and ACP-shaped messages, bridging the agent CLI's stdio through whichever sandbox you picked, handling reconnection, snapshots, and so on.
Solid Agent puts that glue in one place. The protocol is small (ACP + four runtime lifecycle methods + a manifest); the refe
Related Skills
TimmyTest
Local test runner for vibe coding and AI coding agents. Run tests, find likely test gaps, and hand concise fai
Robotics Agent Skills
Agent skills that make AI coding assistants write production-grade robotics software. ROS1, ROS2, design patte
Spectoflow
Agent-agnostic spec-driven development framework with a real-time local dashboard — works with Claude Code, Gi
Slm Setup
Public spec and stdio MCP bridge for a local coding SLM. Premium agents in Cursor, GitHub Copilot, and Claude
Specbridge
Open, model-agnostic spec runtime for existing Kiro projects - bring your .kiro steering and specs to Claude C
Specpipe
Spec-first development toolkit for agentic AI coding agents — skills + guardrails for Claude Code, Codex, Curs
Related Agents
Aux Delegator
Invokes external CLI coding agents (Codex, OpenCode, Claude -p, ACP). Use whenever a task should be offloaded
Claudable
by Ethan Park - Claudable is an open-source web builder that leverages local CLI agents, such as Claude Code a
Kimi Worker
Forward a coding, diagnosis, or review task to Kimi K3 (Moonshot) through the local kimi-worker.sh runtime. Us