Product Overview · June 2026

Your AI agents are making
thousands of decisions.
Who's watching?

TrustLoop is the governance layer between your AI agents and the world. Every action logged, every risk evaluated, every dangerous move stopped — in real time.

<50ms
Governance overhead
1 line
To connect any agent
100%
Tool call visibility
0
Code changes to your agent
The Problem
AI agents act fast.
Mistakes are expensive.
When an AI agent makes a bad decision — sending a sensitive email, deleting the wrong records, approving an unauthorised payment — the damage is done before anyone knows.
No visibility
Your AI agent sends 4,000 emails to the wrong segment. You find out from a customer complaint — two hours later.
No audit trail
The regulator asks "what did your AI do last Tuesday?" You have logs scattered across three systems — none of them complete.
No kill-switch
Your AI agent starts behaving unexpectedly. You have no way to stop a specific action without taking the entire system down.
Capabilities
Everything you need to govern AI at scale
TrustLoop wraps around your AI agents, model API calls, and employee AI usage — without changing how any of them work.
Real-time audit log
Every tool call your AI makes is captured in a tamper-proof log. What was requested, what arguments were passed, what decision was made, and when. Searchable, filterable, exportable.
Compliance
Instant kill-switch
Block any tool across all your agents with one click. No code change, no deployment, no waiting. The block takes effect on the next call — within milliseconds. Unblock just as fast.
Emergency control
Human approval gates
Write rules like "any wire transfer over £10,000 needs approval" in plain English. When the AI triggers the rule, a notification goes to Slack, Teams, or email — the action waits until a human decides.
High-stakes actions
Plain-English governance rules
Set policies without writing code. "Block any action that deletes data." "Require approval before sending emails to more than 100 recipients." TrustLoop's AI evaluates each call against your rules automatically.
Policy engine
Blockchain-anchored audit trail
Log hashes are anchored to the Polygon blockchain every hour. This creates an independently verifiable proof that your audit trail hasn't been tampered with — useful for regulatory submissions and legal disputes.
Tamper-proof
PII & secrets masking
Before any tool call is logged, TrustLoop automatically detects and masks personally identifiable information (names, emails, card numbers) and secrets (API keys, passwords). What's stored is safe by default.
GDPR · CCPA
Multi-channel notifications
Approval requests and alerts land wherever your team works — Slack, Microsoft Teams, Discord, or email. One-click approve or deny directly from the notification. No dashboard login required.
Slack · Teams · Discord
Multi-agent dashboard
One view across all your AI agents. See which agents are most active, which tools they're calling, block rates, risk scores, and pending approvals. Filter by agent, tool, or time period.
Visibility
How It Works
One decision point, before every action
TrustLoop sits between your AI agent and the outside world. Every tool call passes through it — in under 50 milliseconds.
1
AI decides to act
Your agent wants to call a tool — send email, delete record, transfer funds
→
2
TrustLoop intercepts
Call reaches TrustLoop before touching any external system
→
3
Governance check
Kill-switch, approval rules, risk scoring — all applied in <50ms
→
ALLOWED / BLOCKED / PENDING
Decision returned. Action only proceeds if permitted.
ALLOWED
Call passes all checks. Logged and forwarded to your system. Agent proceeds normally.
BLOCKED
Call matches a kill-switch or block rule. Action stopped instantly. Agent receives a clear reason.
PENDING APPROVAL
Call matches an approval rule. Your team is notified. Action waits until a human approves or denies.
Integrations
Works with your existing AI stack
No rip-and-replace. TrustLoop plugs into whatever AI framework you're already using — often with a single environment variable change.
Drop-in Proxy
Change one env var. All OpenAI, Anthropic, or Gemini calls routed through TrustLoop automatically.
Zero code change
Python SDK
pip install trustloop-sdk. Works with LangChain, CrewAI, AutoGen, or any custom agent.
pip install
Node.js SDK
npm install trustloop. TypeScript support. Works with any Node-based agent framework.
npm install
Custom GPT Actions
Import TrustLoop's OpenAPI schema into your GPT. Full governance for any ChatGPT Enterprise deployment.
OpenAI Enterprise
Claude / MCP
Connect Claude Desktop via MCP URL. Every tool call intercepted before Claude executes it.
Anthropic
LangChain
Wrap your tool list with tl.wrap_tools(). One line — every tool governed automatically.
Python
CrewAI
Use @governed_tool() decorator on any CrewAI tool. Governance without touching agent logic.
Python
No-code (n8n / Make)
Add a single HTTP request step before any sensitive action in your workflow. No SDK needed.
REST API
Who It's For
Built for teams deploying AI in production
CTO / Head of Engineering
Ship AI faster with a safety net
  • Deploy agents confidently knowing dangerous actions are blocked
  • Instant kill-switch without a deployment cycle
  • Full observability across all agents in one dashboard
  • Connects in hours, not weeks
Chief Compliance Officer
Answer the regulator's question
  • Complete audit trail: what the AI did, when, and why
  • Blockchain-anchored — independently verifiable
  • GDPR/CCPA-ready: PII masked before storage, DSAR export
  • Data retention controls by plan
CEO / Board
AI with accountability
  • Demonstrates responsible AI deployment to customers and auditors
  • EU AI Act readiness — governance documentation built in
  • Reduces liability from autonomous AI decisions
  • Visible ROI: prevented incidents, compliance costs avoided
Use Cases
Real governance for real risks
Financial Services
Block unauthorised wire transfers
Any payment above a threshold triggers a human approval request to the finance team before the AI can proceed.
Healthcare
Control access to patient records
Audit every access to sensitive records. Block any export that includes PII. Full HIPAA-ready logging.
E-commerce / Marketing
Prevent mass email mistakes
Require approval before any campaign over 1,000 recipients. Stop the wrong-list mistake before it happens.
Legal / Professional Services
Govern document generation
Ensure sensitive documents require review before being sent externally. Full record of what was drafted and when.
SaaS / Product
Safe AI features in production
Ship AI-powered features with governance built in. Kill a misbehaving tool without a deployment. No downtime.
Enterprise IT
Control over AI infrastructure
One governance layer for all AI agents across the business. Consistent policy enforcement regardless of which model or framework each team uses.
Why TrustLoop
Governance infrastructure, production-ready
Every component of AI governance — audit trail, rule engine, approval workflow, PII protection — production-hardened and ready from day one.
Active enforcement, not passive logging
Every AI action is evaluated against your governance rules in real time. Actions are allowed, blocked, or held for approval — before they execute, not after.
Evidence that stands up to scrutiny
Every action is anchored to a public blockchain — tamper-evident and independently verifiable. The standard regulators, auditors, and enterprise procurement teams actually require.
One layer governs your entire AI footprint
Agents, model API calls, employee AI tools — all governed through a single integration. No blind spots, no per-agent configuration. New agents are covered from their first call.
Pricing
Start free. Scale as you grow.
Self-serve plans are month-to-month — upgrade or cancel anytime. Enterprise agreements available with custom data residency, SLAs, and DPA.
Free
£0 /mo
Get started and connect your first agent
  • 5,000 tool calls/month
  • 1 agent
  • 7-day log retention
  • Kill-switch
  • Approval rules
  • Dashboard + audit log
Starter
£29 /mo
Small teams running production agents
  • 100,000 tool calls/month
  • 3 agents
  • 30-day log retention
  • Everything in Free
  • Slack / Teams alerts
  • CSV export
Most Popular
Growth
£149 /mo
Scaling teams with multiple agents
  • 1,000,000 tool calls/month
  • 10 agents
  • 90-day log retention
  • Everything in Starter
  • Blockchain anchoring
  • Compliance export (DSAR)
Enterprise
Custom
Large orgs needing custom data controls
  • Unlimited calls + agents
  • Custom data residency (BYOC)
  • MSA + DPA contract
  • SSO / SAML
  • SOC2 documentation
  • Dedicated SLA + support
Ready to govern your AI agents?
Free to start. Connected in under 2 hours. No credit card required.
Start Free → See How to Connect
Enterprise enquiries: hello@trustloop.live · Technical docs: trustloop.live/technical