# buildAutonomousInstructions

**Kind:** Function

**Source:** [`atloria-monorepo/libs/agent-core/src/autonomous/autonomous-loop.ts`](https://github.com/sherkety/atloria/blob/main/atloria-monorepo/libs/agent-core/src/autonomous/autonomous-loop.ts#L94)

Build the autonomous investigation-verify loop instructions.

This generates a system prompt section — not a complete system prompt.
Domain layers combine it with their own base instructions + tool usage guide.

The prompt teaches the agent to:
1. Investigate before acting (read files, explore structure, check config)
2. Prefer existing tools/scripts over writing from scratch
3. Plan based on what it discovers (not on assumptions)
4. Verify after every significant action
5. Loop until quality gates pass

## Signature

```ts
function buildAutonomousInstructions(options: AutonomousInstructionsOptions): string
```

## Parameters

| Name | Type |
|---|---|
| `options` | `AutonomousInstructionsOptions` |

**Returns:** `string`