# frootai-mcp

Source: https://glimind.com/server/pypi%2Ffrootai-mcp
Registry id: `pypi/frootai-mcp`
Status: unknown

## Install

```json
{
  "mcpServers": {
    "frootai-mcp": {
      "command": "uvx",
      "args": [
        "frootai-mcp"
      ]
    }
  }
}
```

Claude Code:

```sh
claude mcp add frootai-mcp -- uvx frootai-mcp
```

## Tools (40)

- `agent_build`: architecture guidance + recommended play
- `agent_review`: security, quality, compliance audit
- `agent_tune`: production readiness + tuning recommendations
- `compare_models`: side-by-side model comparison
- `compare_plays`: side-by-side play comparison
- `embedding_playground`: v6.7+ similarity comparison with auto-selected backend (azure-openai 1536-dim when env set, else char-ngram TF-IDF cosine, with jaccard back-compat)
- `estimate_cost`: itemized monthly cost per play
- `evaluate_quality`: run quality evaluation (groundedness, relevance, etc.)
- `fetch_azure_docs`: Azure documentation links
- `fetch_external_mcp`: discover MCP servers
- `generate_architecture_diagram`: Mermaid.js diagrams
- `get_architecture_pattern`: architecture decision guides
- `get_azure_pricing`: service pricing by tier
- `get_froot_overview`: complete framework summary
- `get_github_agentic_os`: .github Agentic OS guide
- `get_model_catalog`: model catalog with pricing tiers
- `get_module`: read any module in full
- `get_play_detail`: full play info with infra, tuning, complexity
- `get_play_spec`: SpecKit with WAF alignment
- `get_primitive_detail`: get detail for a specific primitive
- `inspect_wiring`: check what primitives are connected
- `list_community_plays`: community plugin marketplace
- `list_marketplace`: browse AI primitives by type
- `list_modules`: browse FROOT knowledge modules by layer
- `list_solution_plays`: list all 101 plays with filters
- `list_templates`: available scaffold templates by complexity
- `lookup_term`: AI/ML glossary lookup (comprehensive glossary)
- `preview_scaffold`: dry-run preview of scaffold output
- `prompt_dry_run`: test a system prompt with a single sample query without deploying. Calls Azure OpenAI when AZURE_OPENAI_ENDPOINT is set, else uses a deterministic shim (zero to
- `run_eval_live`: POST test cases to a live endpoint, score with simple/LLM-as-judge evaluator, return aggregate mean/p95 vs threshold + per-case rows
- `scaffold_component`: copy-paste-ready code for a single Azure AI
- `scaffold_play`: scaffold a new play with DevKit structure
- `scaffold_status`: check scaffold completeness
- `search_knowledge`: BM25 full-text search across all modules
- `search_marketplace`: keyword search across all primitives
- `semantic_search_plays`: BM25-powered play matching
- `smart_scaffold`: describe what you want, get the best play
- `validate_config`: validate AI config parameters
- `validate_manifest`: validate fai-manifest.json schema
- `wire_play`: generate fai-manifest.json for a play

Measured by Glimind (https://glimind.com): MCP handshake + tool list only, no tool is ever called. CC BY 4.0.