> 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/6.-intent-and-trajectory-governance/6.7-trajectory-summary-state.md).

# 6.7 Trajectory Summary State

Single-step action analysis is insufficient for long-running agents.

Trajectory governance therefore requires a way to summarize the governance-relevant condition of an evolving chain rather than examining only the latest local action in isolation.

The **Trajectory Summary State** provides that higher-level view.

#### Purpose

A trajectory may contain:

* multiple task states,
* several delegated agents,
* changing targets,
* intermediate tool calls,
* verification outcomes,
* external effects,
* and previous admissions.

The governance system needs to retain enough information to answer:

> **What has this trajectory become relative to the intent that originally legitimized it?**

The summary state therefore represents the relevant accumulated condition of the trajectory for continued verification and admission.

#### Local State vs Trajectory State

A local state answers:

```
Is this individual step plausible?
```

A trajectory summary asks:

```
Does the sequence as a whole
still remain inside the authorized trajectory?
```

These may produce different conclusions.

#### Example

Consider:

```
Step 1:
Read configuration
```

```
Step 2:
Compare configuration
```

```
Step 3:
Modify one service
```

```
Step 4:
Modify dependent services
```

```
Step 5:
Modify shared production policy
```

Step 5 may appear locally connected to Step 4.

The trajectory summary may nevertheless reveal that the action chain has crossed the authorized boundary established at Step 1.

#### Public Specification Boundary

Paper 5 contains a fuller formal trajectory state model.

GitBook v2.0 uses **Trajectory Summary State** as the canonical public concept for retaining trajectory-level governance context without reproducing every internal research symbol.

The key requirement is that trajectory governance must not discard prior governance-relevant state merely because the most recent step appears locally acceptable.