> For the complete documentation index, see [llms.txt](https://docs.lpp-minduniverse.org/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.lpp-minduniverse.org/lingua-pactum-protocol-lpp-documentation/5.-lpp-admission-kernel/5.7-delegation-constraints.md).

# 5.7 Delegation Constraints

AI systems increasingly delegate work across agents, tools, organizations, and trust domains.

Delegation creates a central governance risk:

> **Can delegated authority silently become broader than the authority from which it originated?**

LPP's answer is no.

The canonical delegation principles are:

* bounded delegation,
* acyclicity,
* no self-expansion,
* delegatee restrictions.

#### Bounded Delegation

Delegated authority must remain within the authority held by the delegator.

Conceptually:

```
DelegatedScope
⊆
DelegatorScope
```

A principal cannot delegate authority that it does not possess.

#### Acyclicity

Delegation must not create circular authority structures.

The hardened invariant is:

```
¬Cycle(D)
```

where `D` is the delegation graph.

Circular delegation creates ambiguity about authority origin and may allow authority to appear self-reinforcing.

#### No Self-Expansion

Delegation cannot be used to manufacture broader authority.

The following is invalid:

```
Agent A has Scope S
        ↓
delegates to Agent B
        ↓
Agent B returns expanded Scope S+
        ↓
Agent A treats S+ as legitimate
```

Adding an agent does not create new authority.

#### Delegatee Restrictions

Authority may be constrained by who is permitted to receive delegated power.

A legitimate delegation may therefore specify:

* recognized delegatees,
* organizational boundary,
* model or agent restrictions,
* trust domain,
* target systems,
* consequence ceilings,
* or redelegation rules.

#### Incomplete Delegation Chain

If the legitimate delegation chain cannot be reconstructed, the action must not silently proceed.

The formal operational model treats incomplete delegation as a structural failure condition.

#### Delegation and Human Intent

Delegation also creates semantic risk.

Even where the authority chain remains structurally valid, a downstream agent may reinterpret the authorized objective.

This becomes the subject of Chapter 6:

> **Intent & Trajectory Governance**

Authority continuity and intent continuity are related.

They are not identical.