Skip to main content
Glama

Chaperone

Household consent kernel for ambient MCP — pantry/reorder skin for the Amazon Developer Hackathon 2026 Alexa+ track.

Model proposes · policy decides. Irreversible tools use propose_* / commit_* with hash-locked payloads. Demo runs on a simulated Alexa+ host (no device required).

Status

Phase 5 — AmbientBench (100 scenarios, held-out sealed) + AWS DynamoDB/KMS adapters (offline stubs default).

Related MCP server: MCP Scope Consent

Docs

Quick start

pnpm install
pnpm dev:mcp          # terminal 1 — MCP on :3333
pnpm dev:host         # terminal 2 — sim-host on :3000

Open http://127.0.0.1:3000/login

Judge password: household (see apps/sim-host/.env.example)

Demo family: Maya (adult), Leo (child), Guest.

MCP health: http://127.0.0.1:3333/health

AmbientBench (offline)

pnpm eval:validate
pnpm eval:offline

No AWS keys required. See docs/AMBIENTBENCH.md.

MCP Inspector

npx @modelcontextprotocol/inspector

Connect with transport Streamable HTTP to http://127.0.0.1:3333/mcp.

You should see tools hello_chaperone, propose_order, commit_order, list_household and resources ui://chaperone/hello.html, ui://chaperone/confirm.html.

Judge access

  • Sim-host: password household — Maya / Leo / Guest selectors (not biometrics)

  • AmbientBench: pnpm eval:offline (no AWS)

  • AWS dual-write: set CHAPERONE_AWS=1 in apps/mcp-server/.env (see .env.example)

Related MCP Connectors

Related MCP Servers

  • A
    license
    A
    quality
    B
    maintenance
    Default-deny action registry, append-only spend ledger, and human sign-off audit trail (MCP tools).
    6
    MIT
  • A
    license
    Not graded
    quality
    D
    maintenance
    Hosted remote MCP server for managing consent scope receipts, enabling checking, issuing, and auditing consent policies and tool calls.
    MIT
  • F
    license
    Not graded
    quality
    B
    maintenance
    Enables MCP-compatible clients like ChatGPT and Claude to manage a household's shared food inventory, meal plans, recipes, and preferences through natural language, with atomic updates and audit logging.
    -
  • A
    license
    Not graded
    quality
    B
    maintenance
    Enables smart home automation via MCP with human-in-the-loop approval gates, providing device discovery, state queries, policy checks, and proposal-based execution of actions and scenes that only proceed after explicit user confirmation, with full audit logging.
    MIT