CAIN-42 CAIN StudioCAIN-42 invariant · E35
E35-DECISION-REQUIRE_HUMAN: plan-check-act decision REQUIRE_HUMAN is first-class
Held · CAIN-42 Evolution 35 -- Continuous Governance Intelligence Fabric
Last reviewed 2026-10-01
held niche Core guarantees · family plan_check_act
What this rule means
CAIN-42 must always satisfy: plan-check-act decision REQUIRE_HUMAN is first-class. It is one of 799 invariants checked for CAIN-42 Evolution 35 -- Continuous Governance Intelligence Fabric. An invariant is a rule the system may never break, whatever an agent or attacker does; the test suite tries to break it across many scenarios and records the result.
Recorded detail
'REQUIRE_HUMAN'Products this protects
- CAIN Agent Lab — A safe lab to try agent ideas before they touch anything real.
- CAIN Governability Index — A score for how governable an AI system is.
- CAIN Governance API — One API for every governance question about an agent.
- CAIN Governance Cloud — Governance as a managed cloud service.
- CAIN Governance SDK — Libraries for building governance into your own products.
- CAIN Intelligence — Insights across all your agents' governance data.
- CAIN Marketplace — A marketplace for governed agent services.
- CAIN Policy Compiler — Write a rule in plain terms and get an enforceable policy.
- CAIN Proof Exchange — Swap proofs between companies in a standard format.
- CAIN Red Team — Continuous red-teaming of your whole agent setup.
- CAIN Simulator — Simulate agents and rules together to see what would happen.
Verify it in your browser
Your browser downloads the bundle's SHA256SUMS manifest and the file(s) behind this page, hashes them with SHA-256 locally (WebCrypto), and compares. A match shows the record you are reading is the published one; it does not by itself prove who published it (see the signed claims registry and the bundle verifier for that).
Check it yourself
Browse the raw bundle · How to reproduce it · SHA-256 manifest
Scope: An in-process TESTED library; not hosted. The intelligence layer is deliberately NOT in the trusted authorization root; it can only produce recommendations.
Related rules
- E35-RISK-environment: risk dimension environment is separate
- E35-RISK-reversibility: risk dimension reversibility is separate
- E35-FABRIC-recommendation_engine: the intelligence fabric exposes recommendation_engine
- E35-FABRIC-control_optimizer: the intelligence fabric exposes control_optimizer
- E35-FABRIC-failure_learner: the intelligence fabric exposes failure_learner
- E35-FABRIC-evaluation_engine: the intelligence fabric exposes evaluation_engine
- E35-FABRIC-research_graph: the intelligence fabric exposes research_graph
- E35-FABRIC-experiment_engine: the intelligence fabric exposes experiment_engine
← E35-DECISION-ESCALATE · all 799 · E35-DECISION-REAUTHORIZE →
Try CAIN-42 on your own agents
Create a free account and every new account starts with a 7-day trial of the full platform. Or try the sandbox first, with no account at all.
Create a free account → · Try the sandbox · See the whole ecosystem