> ## Documentation Index
> Fetch the complete documentation index at: https://engramviz.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Engram documentation

> Capture memory operations, explain bad agent answers, replay controlled repairs, and preserve the fix as a regression test.

Engram is an open-source memory reliability workspace for AI agents. It helps you answer one concrete engineering question:

> This agent gave a bad answer. Which memory decision caused it, what changes if I correct that decision, and how do I prevent the failure from returning?

<CardGroup cols={2}>
  <Card title="Capture your first turn" icon="radio-tower" href="/quickstart">
    Install the CLI and SDK, run your agent with local capture, and inspect the resulting turn.
  </Card>

  <Card title="Understand the lifecycle" icon="workflow" href="/concepts/memory-lifecycle">
    Learn the difference between storing, retrieving, loading, updating, and consolidating memory.
  </Card>

  <Card title="Diagnose an incident" icon="triangle-alert" href="/investigate/incidents">
    Connect a bad answer to the exact memory evidence available at each stage.
  </Card>

  <Card title="Run the reference failure" icon="flask-conical" href="/examples/stale-location">
    Reproduce a stale-location answer, repair it on a branch, and execute the exported regression.
  </Card>
</CardGroup>

## The core loop

<Steps>
  <Step title="Observe">
    Capture what the memory system stored, updated, considered, selected, and loaded into model context.
  </Step>

  <Step title="Diagnose">
    Find the earliest recorded memory decision that can explain the bad answer without inventing missing evidence.
  </Step>

  <Step title="Intervene">
    Change memory state, eligibility, ranking, selection, or context on an isolated branch while preserving the source run.
  </Step>

  <Step title="Replay">
    Reproduce the baseline, rerun every supported stage, and compare the first observable divergence.
  </Step>

  <Step title="Prove">
    Export semantic lifecycle and answer assertions, then run them across paraphrases, entity changes, score ties, and distractors.
  </Step>
</Steps>

## Start locally

```bash theme={"dark"}
npm install --save-dev @engramviz/cli
npm install @engramviz/sdk
npx --yes @engramviz/cli init --project my-agent
npx --yes @engramviz/cli dev
```

Then run your agent with the capture environment injected:

```bash theme={"dark"}
npx --yes @engramviz/cli run -- npm run my-agent
```

Engram Studio opens locally at `http://localhost:3100/?mode=incidents`.

<Info>
  Engram visualizes observable application behavior, not hidden chain-of-thought or model activations. A loaded memory was available to the model; availability alone does not prove that the model relied on it.
</Info>

## How the pieces connect

```mermaid theme={"dark"}
flowchart LR
  A[Agent and memory provider] --> B[Engram SDK or adapter]
  B --> C[Memory Telemetry v2]
  B --> D[Turn Envelope v1]
  C --> E[Local append-only capture]
  D --> E
  E --> F[Engram Studio]
  F --> G[Incident diagnosis and intervention]
  G --> H[Capability-declared policy replay]
  H --> I[Semantic regression matrix]
```

The 3D brain is a synchronized educational and evidence map. The Incident workspace is the engineering surface for diagnosis, intervention, replay, and verification.