data-ai MCP Server
A provider-agnostic scaffolding kit for running structured multi-agent workflows in your codebase.
Discovered via github-topic:mcp and last synced 3mo ago.
1. Install the package
npx @cardor/agent-harness-kit
2. Add to claude_desktop_config.json
{
"mcpServers": {
"-cardor-agent-harness-kit": {
"command": "npx",
"args": [
"@cardor/agent-harness-kit"
]
}
}
}Config file location: ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) / %APPDATA%\Claude\claude_desktop_config.json (Windows)
lead
Mark an acceptance criterion as met. Criterion IDs come from `tasks.acceptance_get`
Search the `docsPath` folder for content matching the query
What it shows
SQLite
Edit it?
Commit?
cjs} ├── AGENTS.md ├── health.sh ├── .harness/ └── .codex/ ├── config.toml ← MCP server registration └── agents/ ├── lead.toml ├── explorer.toml ├── builder.toml ├── reviewer.toml └── default.toml ← overrides Codex's built-in default agent → routes to lead ``` ### What each file does
Returns all acceptance criteria for a task with their `id`, `task_id`, `criterion` text, and `met` status. Use the returned `id` values with `tasks.acceptance.update`
builder
Responsibility