Kevros — Governance-as-a-Service for AI Agents

Every agent action verified. Every decision recorded. Every record signed.

What This Does

Kevros adds cryptographic governance to any AI agent. Before your agent takes an action (deploy code, move money, access data), Kevros evaluates it against your policies and returns a signed decision: ALLOW, CLAMP (adjusted to safe bounds), or DENY.

Every decision is appended to a hash-chained, tamper-evident evidence ledger. Auditors can verify the entire chain without your source code.

Quick Start

from kevros_governance import GovernanceClient

client = GovernanceClient(agent_id="my-agent")
result = client.verify(
    action_type="trade",
    action_payload={"symbol": "AAPL", "qty": 100},
    agent_id="my-agent",
)
print(result.decision)  # ALLOW, CLAMP, or DENY

Or use the API directly:

# Get a free API key (instant, no credit card)
curl -X POST https://governance.taskhawktech.com/signup 
  -H "Content-Type: application/json" 
  -d '{"agent_id": "my-agent"}'

# Verify an action
curl -X POST https://governance.taskhawktech.com/governance/verify 
  -H "X-API-Key: kvrs_..." 
  -H "Content-Type: application/json" 
  -d '{"action_type": "deploy", "action_payload": {"service": "api"}, "agent_id": "my-agent"}'

Governance Primitives

Primitive	What It Proves	Cost
Verify	Agent is authorized — signed ALLOW/CLAMP/DENY	$0.01
Attest	Action happened — hash-chained evidence	$0.02
Bind	Intent matched command — cryptographic binding	$0.02
Bundle	Compliance evidence package — independently verifiable	$0.25
Media Attest	Media file integrity — SHA-256 in provenance chain	$0.05

Free endpoints: verify-outcome, verify-token, verify-certificate, reputation lookup, passport, media verify.

Payment

Free tier: 1,000 calls/month, instant signup, no credit card
x402 (USDC on Base): Pay per call, no API key needed
Subscription: Scout $29/mo, Sentinel $149/mo, Sovereign $499/mo

Why Agents Need This

Audit trails: When regulators ask "who authorized this agent action?", you have cryptographic proof
Fail-closed safety: If governance fails, the agent stops. Not the other way around
Trust between agents: Agent B can verify Agent A's release token without trusting Agent A
Evidence chain: Hash-chained, append-only, independently verifiable

Integration

Python SDK: pip install kevros
TypeScript SDK: npm install @kevros/agentkit
MCP: https://governance.taskhawktech.com/mcp/
A2A: https://governance.taskhawktech.com/.well-known/agent.json
x402: https://governance.taskhawktech.com/.well-known/x402
REST API: https://governance.taskhawktech.com/api

Works with LangChain, CrewAI, OpenAI Agents SDK, Microsoft Agent Framework, and any HTTP client.

kevros-governance

Installation