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

# Agent Composer Templates

> Agent Composer templates: Basic Search, Agentic Search, and Enterprise templates like Device Log Analysis

## Overview

Agent Composer offers pre-built **templates** so you can get started quickly. Availability is by template type:

* **Basic Search** and **Agentic Search** — available to **self-serve** users.
* **All other templates** in the [Template Catalog](/examples/templates-catalog) (e.g. Device Log Analysis, Deep Research, domain-specific templates) are available only for **Enterprise**. There are no other Enterprise-only features—only access to these additional templates. [Request a demo](https://contextual.ai/request-a-demo) to use them.

| Template                | Availability | Best for                                                                                          |
| ----------------------- | ------------ | ------------------------------------------------------------------------------------------------- |
| **Basic Search**        | Self-serve   | Quick lookups, simple Q\&A, direct search-to-response                                             |
| **Agentic Search**      | Self-serve   | Multi-step retrieval, iterative research, complex Q\&A                                            |
| **Device Log Analysis** | Enterprise   | Automated root cause analysis of device logs (e.g. 3GPP LTE), multi-agent parsing and RCA reports |

***

## Basic Search

**Basic Search** (also referred to as Fast Search in the template catalog) is a streamlined, lightweight template for quick document retrieval and response generation.

* **Simple architecture:** Direct search-to-response pipeline without multi-turn research
* **Fast performance:** Minimal configuration for rapid queries
* **Basic RAG:** Standard retrieval with citation and groundedness tracking

**Use cases:** Quick lookups, simple Q\&A, straightforward document search.

For more templates in this style (e.g. Fast Search YAML), see the [Template Catalog](/examples/templates-catalog#enterprise-&-legal).

***

## Agentic Search

**Agentic Search** is a multi-step research template. The agent uses iterative retrieval and tool use to answer complex questions with citations.

* **Multi-turn research:** Plans searches, refines queries, and synthesizes answers from multiple retrievals
* **Structured agent loop:** Uses defined tools (e.g. search over datastores) with guardrails
* **Citations:** Grounded responses with source references

**Use cases:** Technical documentation Q\&A, research over internal docs, customer support with deep lookup.

**Try the demos:** All of these agents use the Agentic Search template (or the same multi-step research pattern). Launch the live demo or read the example page.

* [Rocket Science](/examples/rocket_science) — [Launch](https://demo.contextual.ai/propulsion-agent): Root cause analysis for rocket engine anomalies
* [Raspberry Pi Agent](/examples/raspberry-pi-agent) — [Launch](https://demo.contextual.ai/customer-engineering): Technical support for Raspberry Pi hardware
* [Material Science Agent](/examples/material-science-agent) — [Launch](https://demo.contextual.ai/material-science): Research assistant over materials papers (ELI5 style)
* [Test Program Generation](/examples/test-automation) — [Launch](https://demo.contextual.ai/test-code-generation): Generate OpenTAP test programs and drivers
* [3GPP Spec Explorer](/examples/3gpp-spec-explorer) — [Launch](https://demo.contextual.ai/3gpp-agentic-search): Search and reason over 3GPP specifications (Release 19 and others)

**Guides:** [Template Catalog](/examples/templates-catalog) for related templates.

***

## Device Log Analysis (Enterprise)

The **Device Log Analysis** template is available in Agent Composer for Enterprise customers. It automates root cause analysis (RCA) on device failure logs (e.g. 3GPP LTE eNB/UE logs) by parsing logs, building searchable databases, and running multi-agent analysis to produce timelines, visuals, and detailed RCA reports.

[Example page](/examples/device-log-analysis) · [Launch demo](https://demo.contextual.ai/log-analysis)

**In production:** You upload a log file and optionally a debug reference guide, then ask a query (e.g. *"Why did the call drop?"*). The rest is automatic: the system parses logs, builds databases, runs root cause analysis, can generate Python scripts that run in a secure VM, and produces visuals and a detailed RCA report. This is a longer-running, multi-agent workflow aimed at complex, routine knowledge work.

**Next steps:** [Request a demo](https://contextual.ai/request-a-demo) to get access.

**Docs and demo:** The [example page](/examples/device-log-analysis) has the full description, example outputs (causation timelines, system health visuals), and report structure. The live demo includes **three** preloaded sample datasets: **3GPP Wireless**, **ATE Test Equipment**, and **Mazda Infotainment** (each: log + debug reference). Try it at [demo.contextual.ai/log-analysis](https://demo.contextual.ai/log-analysis).