BETA Design partner program — fintech, insurtech, healthtech teams shipping AI agents into production. Apply →
OPEN SOURCE · PRE-EXECUTION ACTION AUTHORIZATION

Action authorization layer
for AI agents.

Permit0 is the action authorization layer for AI agents — the deterministic policy enforcement between what an agent decides and what it executes. We provide a canonical taxonomy of agent actions across 20 domains, risk defaults, and compliance mappings. Built for engineering teams shipping agents into production with trust.

support_agent.py
from permit0 import guard, permit

@guard
def refund_customer(charge_id, amount):
    permit.require("payments.refund", {
        "amount": amount,
        "actor": "support_agent_v2",
    })
    return stripe.Refund.create(charge_id)

# agent fires this after a customer complaint
refund_customer("ch_3O9k...", amount=85000)
✗ BLOCKED  action=payments.refund
risk: 82/100  flags: FINANCIAL · MUTATION
reason: amount > $1,000 requires human approval
escalated: → human_review
119
Canonical actions
20
Action domains
9 × 10
Risk dims (flags × amplifiers)
<5ms
p99 enforcement
3
Deployment modes
10+
AI compliance mappings
Compatible with the agent stack you already run
LangChain Model Context Protocol CrewAI OpenClaw AutoGen LangGraph OpenAI Agents SDK Claude Skills
Why permit0

The layer no incumbent will build.

Identity layers control which agents exist. Observability layers report what already happened. Platform vendors govern only what runs inside their boundary. Cross-platform action authorization — the layer that decides whether a specific action gets to execute, in real time, before it does — is structurally opposed to every incumbent's business model. So we built it.

Engineering credibility you can audit

Built by engineers who shipped service infrastructure at Google Cloud and enterprise audit and trust at Salesforce. The architecture is published, the code is open, the risk model is mathematically defensible.

Deterministic where it counts

The decision path is rules and cryptography. No model sits in the hot path. Every block, allow, and escalation is replayable, signed, and grounded in a policy your team owns. AI helps author the policy. It never makes the call.

Cross-platform by design

Salesforce Agentforce governs Salesforce. Microsoft Agent 365 governs Microsoft. Okta Agent Gateway governs identity. Permit0 governs the action — whether the agent reads from Salesforce, processes through OpenAI, and refunds through Stripe. One policy, one audit trail, one source of truth across every framework, every tool, every cloud.

Built for four roles

Four roles. One control plane.

Every policy engine ships an empty policy. Permit0 is the layer that ends the patching — providing central policy control, compliance-grade auditing, and the trust your team needs to deploy agents into production.

Developers

Ship agents faster, with authorization built in.

Stop writing permission checks inside agent code. Full risk taxonomy and production-grade defaults — pip install, ship.

  • 5-minute integration with LangChain, MCP, CrewAI, OpenClaw, and raw Python tools
  • 119 default actions across 20 domains, with risk policies on day one
  • Free forever for OSS — full SDK, no rate limits
Get started in 5 minutes
Platform Engineering Leads

One policy layer across every agent your teams ship.

Every framework, every tool, every team — each reinventing authorization. Permit0 is the single control plane. Write the policy once. Enforce it everywhere.

  • Policy portability — the Tool Action Compiler normalizes any tool call into a canonical action across frameworks
  • Session-aware chain detection — dangerous action sequences get caught and escalated
  • VPC-native — cloud, self-hosted, or managed VPC; runs where your data lives
See the architecture
CISOs

Agents do more, without expanding your threat surface.

AI agents are your newest employees — no onboarding, no permission boundaries. 88% of organizations report AI agent incidents. Permit0 puts a deterministic gate between every decision and its execution.

  • Pre-execution enforcement — block bad actions before they fire
  • Cryptographic capability tokens — prompt injection can't bypass a signed gate
  • Global kill switch — your team retains override authority, every deployment mode
Read the security architecture
Compliance Leads

Audit-ready evidence, before the auditor asks.

Compliance has gone from quarterly review to deployment blocker. Permit0 gives your team replayable, exportable evidence — pre-mapped to the frameworks auditors already use.

  • Audit-of-record per decision — every block, allow, and escalation cryptographically signed
  • Replayable evidence — export any agent action in audit-grade format
  • Compliance packs — EU AI Act, Colorado AI Act, NIST AI RMF, ISO 42001, SOC 2, FINRA, HIPAA, SR 11-7, and more
Read the compliance brief
The Permit0 platform

Three components.
One control plane.

Open source at the core. Self-hosted, managed VPC, or cloud — the policy decision runs where your data lives.

Permit0 Engine

The deterministic policy engine. The Tool Action Compiler normalizes any tool call into a canonical action. The Risk Engine scores against 9 flags and 10 amplifiers. Two-stage review pairs a deterministic scorer with an optional LLM reviewer that can only deny or escalate — never approve. Capability tokens bind approved actions to payload hashes, constraint envelopes, and decision provenance.

Architecture spec

Permit0 Taxonomy

A canonical taxonomy of 119 agent actions across 20 domains — each scored across 9 risk dimensions with opinionated defaults. Open, versioned, citable. The vocabulary your auditor can verify against and the substrate every Permit0 deployment runs on. Ship a policy-guarded agent in days, not quarters.

Read the spec

Permit0 Compliance Center

Audit-of-record evidence designed for the frameworks regulated buyers already work with — EU AI Act, Colorado AI Act, NIST AI RMF, ISO 42001, SOC 2, FINRA, HIPAA, SR 11-7, and more. Built to be the certifying authority your auditor recognizes.

Compliance roadmap
Engineered to be referenced

Built to be cited by auditors,
not just adopted by developers.

The taxonomy, the engine, and the deployment model are engineered for the auditor's spreadsheet — not just the developer's IDE.

119
Default canonical actions shipped
20
Action domains covered out of the box
<5ms
p99 policy enforcement latency
9 × 10
Risk flags × amplifiers in the model
10+
Compliance frameworks mapped
Where it matters first

Built for the verticals where one bad action is a regulatory event.

Permit0 leads with the segments that need pre-execution evidence — not post-hoc detection. The 2026 regulatory clock is short.

Fintech & Payments

Agents touching money have the highest blast radius. Refunds, transfers, ledger writes, KYC dispositions, SAR drafting. SR 11-7 and FFIEC examiners increasingly ask for pre-execution evidence on every autonomous action.

EU AI Act Annex III · Colorado AI Act · CFPB Circular 2023-03 · ECOA · SR 11-7 · FFIEC · NACHA · BSA/AML · NYDFS Part 500

Collections & Loan Servicing

Voice and SMS agents under FDCPA, Reg F, and TCPA. Every outbound contact is a 7-in-7 / consent / time-of-day decision the CFPB can subpoena. Bureau-furnishing writes are irreversible under FCRA § 623.

FDCPA · Reg F · TCPA · FCRA · UDAAP · CFPB AI guidance

Insurance & Claims

Claim approve / deny / pay agents. EU AI Act Annex III explicitly lists insurance pricing and claims as high-risk. Denials carry UDAAP and bad-faith exposure under state DOI rules.

NAIC AI Model Bulletin (24+ states) · EU AI Act Annex III · NY DFS Letter 2024-7 · CO SB21-169 · NAIC · State DOI

Healthcare

PHI access, clinical-note generation, patient-facing communication, prior-auth and claims agents. Every action touching a patient record is a HIPAA event. EU AI Act Annex III explicitly lists health AI as high-risk and requires deployment-time controls Permit0 enforces.

HHS Section 1557 (AI nondiscrimination) · HTI-1 / ONC DSI rule · EU AI Act Annex III · HIPAA · HITECH · 42 CFR Part 2

HR & Employment

Resume screening, candidate scoring, performance assessment, benefits enrollment. EEOC has signaled AI hiring tools are subject to existing anti-discrimination law. NYC LL 144 mandates pre-deployment bias audits. Illinois AIVID and Colorado AI Act add state-level disclosure and impact-assessment requirements.

EU AI Act Annex III · EEOC · NYC LL 144 · IL AIVID · CO AI Act · GDPR Art. 22
+ Cross-vertical AI governance
Every Permit0 deployment also maps to the horizontal AI frameworks.
These laws and standards apply across all verticals — establishing the baseline AI governance every regulated industry shares, on top of the vertical-specific provisions above.
EU AI Act Horizontal AI law; Annex III enumerates the vertical-specific high-risk provisions. Effective Aug 2026.
Colorado AI Act Consequential decisions across finance, health, employment, insurance, education. Effective Feb 2026.
NIST AI RMF US federal AI governance reference (AI 100-1), including the GenAI Profile (AI 600-1).
ISO 42001 International AI Management System standard (ISO/IEC 42001:2023).
SOC 2 Type II Security and trust controls — enterprise procurement table stakes.
LIMITED COHORT · OPEN NOW

Design partner program.

Free for high-growth fintech, insurtech, and healthtech teams shipping AI agents into production. Co-develop a vertical compliance pack with the founding team.

Survive the EU AI Act (August 2026) and Colorado AI Act (June 2026) without a retrofit.

Selection criteria Operating in a regulated vertical · Shipping agents that touch money, identity, or PII · Willing to commit a 30-minute weekly working session for 4 weeks.
Resources

The references behind the engine.

Everything Permit0 enforces is documented, versioned, and verifiable.

$ pip install permit0

Action authorization for AI agents.

Open source. Production-grade. Ready for the regulated verticals where post-hoc audit is no longer enough.