# sandbase-harness

> sandbase-harness — sandbaseai-sandbase-harness. Use this tool when you need to deploy and manage AI agents in a secure, self-hosted environment with features like sandboxed sessions and audit/replay capabilities. It solves problems related to AI agent runtime management, security, and transparency, and accepts inputs like git repositories and AI models. Outputs include auditable session logs and replayable interactions, making it ideal for development, testing, and production environments.

Canonical page: https://skillsregistry.net/skills/sandbaseai-sandbase-harness  
JSON: https://api.skillsregistry.net/v1/skills/sandbaseai-sandbase-harness

## Description

Local-first, self-hosted AI agent runtime and MCP bridge with sandboxed sessions, memory, credentials, audit/replay, and a local Console.

## Trust

- **Trust score (0–1):** 0.50
- **Verification tier:** unverified

## Facts

- **Version:** 1.0.0
- **Skill type:** atomic
- **Execution layer:** container
- **Runtime environment:** vm
- **Updated:** 2026-09-28

## Source

- **Source listing:** [GitHub](https://github.com/sandbaseai/sandbase-harness)

## Use it

Resolve this record through the SkillsRegistry MCP server (no auth, read-only):

```
claude mcp add --transport http --scope user skillsregistry https://api.skillsregistry.net/mcp
```

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "get_skill",
    "arguments": {
      "slug": "sandbaseai-sandbase-harness"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/sandbaseai-sandbase-harness` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/sandbaseai-sandbase-harness/pull`

---
SkillsRegistry indexes agent skills from public registries and GitHub. Skills we have analysed are scanned with Circle-IR and scored on six dimensions; each listing states its scan coverage. More: https://skillsregistry.net/llms.txt