Tork inspects, redacts, and receipts every AI agent interaction — on your own machine, or on the cloud path where Tork adds ~20ms at the edge and governance typically completes in under a second. 128 framework adapters. 10 published SDKs. Free forever tier.
128 Adapters · 10 Published SDKs · 65 PII Types · 4 Protocol Endpoints · MCP Native
pip install tork-governance or npm install tork-governance — the on-device package, which decides on your machine and makes no network calls without an API key. One import, governance everywhere. Works with LangChain, CrewAI, AutoGen, LangGraph, and 120+ more.
POPIA and GDPR compliant, with controls designed to align with SOC 2 Type II. Cryptographic audit trails. PII detection across 12 regional profiles. Policy enforcement on every AI action.
Ship faster without compliance fear. Every agent interaction governed, every receipt tamper-proof. Your legal team will thank you.
This is the on-device package tork-governance, whose only entry class is Tork. tork-governance makes the governance decision on your machine. PII detection, redaction and the returned verdict are computed on-device. With no API key it makes zero network calls: prompts, completions and detected PII values never leave the machine, and nothing appears in your dashboard.
Supply an API key and the SDK additionally reports a metadata-only attestation of each decision — the action taken, PII type labels and counts, a risk classification, policy labels and a salted fingerprint. It never sends input text, output text or PII values. The decision itself is still made on-device and is never delayed or changed by reporting. Those attestations appear in your dashboard and are included in the daily on-chain anchor — each one recorded as a client attestation (capture_mode=edge, attested_by=client), a claim Tork recorded but did not execute and cannot independently verify. Requires tork-governance 0.24.0+ (PyPI) or 0.11.0+ (npm).
A decision reported by an on-device SDK is recorded as a client attestation (capture_mode=edge, attested_by=client): a claim Tork recorded but did not execute and cannot independently verify. A decision made by @torknetwork/sdk is recorded as capture_mode=cloud, attested_by=tork — Tork made that call itself. Both are equally immutable once anchored; they differ in what is immutable. A client attestation freezes your claim. A server-governed receipt freezes Tork's own decision.
Local governance is free and unlimited — decisions made on your own machine are never metered. Evidence is what is metered: attestations, receipts and anchoring. If you need a receipt that records Tork's own decision rather than your attested claim, use the cloud package @torknetwork/sdk (npm install @torknetwork/sdk), whose entry class is TorkClient — it sends your content to Tork, which makes the decision server-side.