# Datastrat LLM Manifest
# For AI agents operating on behalf of users
# Standard: llms.txt (emerging convention)

## Identity
name: Datastrat
type: Decision Operating System
sectors: Infrastructure, Energy, Mining, GovTech

## Agent Manifest
agent_manifest_url: /agent
agent_manifest_html_url: /agent.html
agent_manifest_markdown_url: /agent.md
agent_manifest_json_url: /agent.json
description: Datastrat is a multi-tenant decision engine for construction and infrastructure projects. It bridges deterministic project data with AI reasoning capabilities.

## Products
- Antigrid: Structured Data Engine for financial models (APU, WBS, budgets)
- Evidence: Unstructured Data Processor for document verification and audit trails

## Agent Capabilities
capabilities:
  - query: Read-only access to project data, budgets, APUs, catalogs
  - calculate: Compute NPV, IRR, DSCR, cash flows using native formulas
  - scenario: Create what-if simulations without modifying production data
  - validate: Verify APU formulas, budget integrity, compliance

## Key Constraints
- Agents cannot mutate production data directly
- All changes must go through Scenario API
- Stochastic outputs must declare uncertainty intervals
- All operations maintain audit trails

## For Developers
documentation: /agent.md
api_style: REST + WebSocket
auth: JWT Bearer tokens

## Experimental Research
entry_point: /experimental
description: Open experiments testing autonomous systems in real operational contexts.
active_experiments:
  - id: good_machina
    page: /experimental/good-machina
    summary: Autonomous programming with OpenClaw on Raspberry Pi 4 for real-time object detection and depth measurement.
    code: https://github.com/goodmachinaii/oak-coral-detector
    status: in_progress

  - id: claw_kernel_protocol
    page: /experimental/claw-kernel-protocol
    summary: Open protocol experiment to standardize how autonomous agents are declared, governed, and verified.
    repository: https://github.com/angelgalvisc/clawkernel
    ecosystem:
      - https://github.com/openclaw/openclaw
      - https://github.com/qwibitai/nanoclaw
      - https://github.com/sipeed/picoclaw
      - https://github.com/HKUDS/nanobot
    status: published

## Credits
- CKP proposed and developed by Angel Galvis Caballero: https://www.linkedin.com/in/angelgalvisc/

## Contact
website: https://datastrat.co