# Configuration Optimization (/docs/verdent-manager/configuration/configuration-optimization)

> Quick reference for picking the right model and execution mode for each task



A concise cheat sheet for matching Verdent Manager's model and execution mode to the task at hand. For deeper configuration, see the dedicated pages for [Subagents](/docs/verdent-manager/configuration/subagents), [BYOK](/docs/verdent-manager/configuration/byok), and [BYOA](/docs/verdent-manager/configuration/byoa).

***

## Model Selection [#model-selection]

| Task Type                            | Recommended Model | Why                                    |
| ------------------------------------ | ----------------- | -------------------------------------- |
| Quick fixes, small edits             | Claude Haiku 4.5  | Fastest, lowest credit cost            |
| General development                  | Claude Sonnet 4.6 | Best speed/quality tradeoff            |
| Complex architecture, deep reasoning | Claude Opus 4.6   | Strongest reasoning                    |
| Parallel routine tasks               | Claude Haiku 4.5  | Resource-efficient across many Workers |

<Tip>
  Leave **Auto** mode on in the input box and Verdent picks the model for you based on task signals. Disable Auto only when you want to pin a specific model.
</Tip>

***

## Execution Mode Selection [#execution-mode-selection]

| Scenario                               | Mode                   | Why                                    |
| -------------------------------------- | ---------------------- | -------------------------------------- |
| Clear, well-defined task               | Agent Mode             | Direct execution, no planning overhead |
| Unfamiliar codebase, exploration       | Plan Mode              | Review approach before acting          |
| Production changes, architectural work | Plan Mode → Agent Mode | Validate plan, then execute            |
| Fast iteration loops                   | Agent Mode             | Minimal friction                       |

See [Agent Mode](/docs/verdent-manager/advanced-features/agent-mode) and [Plan Mode](/docs/verdent-manager/advanced-features/plan-mode) for full details.

***

## Context Management [#context-management]

For large projects, keep the main context window lean:

* **Delegate to subagents** — `@Verifier`, `@Fast Context`, and custom subagents run in their own context; only their results return to the main agent.
* **Reference files explicitly** — Use `@filename` to pull only what's needed instead of loading whole directories.
* **Chunk large tasks** — Start a new session at natural boundaries (per feature, per module).
* **Plan Mode first** — Validate the approach before burning execution context on the wrong path.

***

## Configuration Priority [#configuration-priority]

When rules conflict, Verdent applies this priority order:

1. **Project Rules** (`AGENTS.md` in project root) — highest priority
2. **Workspace settings** — application workspace-level settings
3. **User Rules** (`~/.verdent/VERDENT.md`) — global defaults
4. **Verdent defaults** — lowest priority

See [Rules](/docs/verdent-manager/configuration/rules) for writing effective rules.

***

## See Also [#see-also]

<CardGroup cols="2">
  <Card title="Subagents" icon="robot" href="/docs/verdent-manager/configuration/subagents">
    Delegate work with isolated context windows
  </Card>

  <Card title="BYOK" icon="key" href="/docs/verdent-manager/configuration/byok">
    Use your own provider API keys
  </Card>

  <Card title="BYOA" icon="plug" href="/docs/verdent-manager/configuration/byoa">
    Run Workers on external Agent runtimes
  </Card>

  <Card title="Rules" icon="list-check" href="/docs/verdent-manager/configuration/rules">
    Project and user rule files
  </Card>
</CardGroup>