Documentation Center

Searchable, versioned docs with table of contents and code highlighting.

v2.0

Getting Started

Platform overview
Quickstart
Running an agent
Reading evaluation reports

Agent Authoring

Prompt structure
Inputs & outputs contract
Evidence rules (VERIFIED/INFERRED)
Verification checklist

Evaluation

Scoring matrix
Risk levels
Verification logs
Agent vs Verified split

Infrastructure

Reproducible env (mise)
CI pipeline
Docker Compose
Terraform

# Quickstart — run any agent locally
make bootstrap          # pin runtimes, install deps, generate .env, test
make test               # full suite (85 tests across Python/Node/Rust)

# Copy a prompt and run it in your coding agent
#   Agents → pick an agent → Copy Prompt → paste into Claude Code / Cursor