F3 · AgentSea guided build

AgentSea
in action.

Four real demonstrations. One pattern: describe the outcome, give the agent the right skills and context, then verify the work.

Observed live10–12 min instructor setupAgentSea (Beta)

01 · The demo story

Orient. Build. Apply.

The platform disappears into the background. The useful part is the repeatable workflow.

01 · ORIENT

Find your way

Accept the terms, take the guided tour, find your private agents and discover what colleagues have shared.

Where things live

02 · BUILD

Describe the goal

Natural language becomes an editable system prompt, selected skills, model configuration and starter prompts.

No code needed

03 · APPLY

Do real work

Transform a rough keynote or analyse an RFP workbook — then inspect, refine and download the result.

Then verify

02 · Orient — first login

Your first 60 seconds in AgentSea

agentsea.synapxe-ai.sg/agents

AgentSea

My Agents

Discover

Recent chats

Agent Builder

Preview — Executive Deck Builder

Chat with AWS Healthcare Keynote…

Settings & Help

DiscoveryAgentSea (Beta)

Create powerful
AI agents

DESIGN • BUILD • SHARE

+ Create AgentDiscover CommunityFAQ & User Guide

MY AGENTS

📊Executive Deck Builder

Creates polished PowerPoint presentations tailored…

~~DRAFT🔒 PRIVATE~~

🏥AWS Healthcare Keynote

Creates professional keynote presentation decks…

~~DRAFT🔒 PRIVATE~~

STEP 1 OF 7

Create your agent

Start here to build your first agent — no code needed.

SkipNext

Accept the terms

Standard responsible-use conditions for AI applications.

Follow the tour

Seven guided steps point out Create, Discover and navigation.

Choose your path

Create a new agent — or start from one a colleague shared.

Interactive recreation of the AgentSea landing page · video 2489

03 · Orient — two surfaces + built-in help

Build privately. Reuse collectively.

0 public agents observed

A visible signal of early internal reuse across public healthcare.

My Agents · Private

Your working space

Create, test and refine before deciding whether anything should be shared.

Discover · Shared

Save a copy

Browse agents colleagues built and adapt a copy for your context. Already live:

NUHS ERM NavigatorKTPH@Home Roster GenieSRE Report Generator

AgentSea Lifeguard — ask for how-to help, give feedback or raise a ticket, in natural language. Shortcut: Ctrl/Cmd+M.

04 · Build — natural-language creation

Describe the outcome. AgentSea builds the configuration.

agentsea.synapxe-ai.sg/agents/create

CREATE AN AGENTSave Agent

💬Create

CreateConfigure

DESCRIPTION

Creates professional PowerPoint presentations tailored for executive audiences.

SYSTEM PROMPT

You are a professional presentation assistant for executive-level PowerPoint decks. First confirm the topic, audience and key points, then use Document Builder to produce a polished PPTX…

987 / 30000

MODEL

Claude Sonnet 4.5▾

SKILLS ✓ auto-selected

Document Builder

Generate PDF, DOCX, PPTX, XLSX from scratch

Document Insights

Microsoft Exchange

TEST AGENT RESPONSE

📊

PowerPoint Creator

Creates professional PowerPoint presentations for executive audiences.

Create a quarterly business review deck

Generate slides for a strategy meeting

Message PowerPoint Creator…Max 30 MB

System prompt

An editable role, workflow and set of instructions.

Model

Claude Sonnet 4.5 — the Bedrock engine underneath.

Skills

Document Builder auto-selected for a deck job.

Everything editable

Business users describe; power users refine.

Recreation of the Configure tab · video 2490

05 · Build — one platform, two depths

Start simple. Go deeper when you need control.

Create

Business users

Describe the goal in plain language
Let AgentSea propose the setup
Test the agent in the same screen

same agent

Configure

Power users

Edit the full system prompt
Choose skills and privacy
Tune suggested prompts and outputs

Today’s bridge: we already wrote the Guideline Assistant system prompt in F2. In F3, we put it into Configure, add Document Analysis, then test it against a synthetic guideline.

06 · Apply — use case A · Document Builder

From rough content to an on-brand executive slide

Before · raw content

The Next Frontier of Human + AI

Human judgement. Agentic execution. System learning.

Human owns

Clinical judgement · Context · Ethical decisions · Accountability

AI provides

Research · Analysis · Documentation · Coordination · Execution support

Healthcare system benefits

Shared learning · Faster innovation · Consistent standards · Population-scale impact

Key statement: the future of public healthcare is not autonomous AI. It is trusted AI working alongside healthcare professionals, within shared governance, platforms and data foundations.

After · on-brand slide

Human-plus-AI in practice

From AI architecture to everyday healthcare work

We've built the architecture. The next step is helping every healthcare professional safely use AI — without needing to understand models, infrastructure or prompt engineering.

Access

Multiple frontier models
Best model for each task
Unified experience

Govern

Enterprise guardrails
Security & compliance
Auditability & oversight

Scale

Shared prompts & workflows
Reusable patterns
Organisation-wide learning

One common Human+AI workspace where innovation can scale safely.

Recreation — AgentSea matched the base deck's style, tone and layout (~80% of the work); the presenter did the final polish · video 2492

07 · Apply — use case B · Excel analysis

Turn rows of supplier responses into a decision-ready view

X Event Content.xlsx — RFP responses

RFP question	SP Sus Office	SP Transf.	Si2
Company registration number	✓	✓	✓
Scope 1 & 2 emissions reported?	Yes	Yes	No
Years in operation	12	8	5
Net-zero commitment + target year?	2040	2045	No
Standard payment terms (days)	30	45	60
Renewable energy in operations?	Yes	Part	No
Support SLA tier	Gold	Silver	Bronze

Overall Sustainability Performance

Category score · higher is better

0.0%

SP Sus Office

SP Transformation

Si2 Pte Ltd

Filter

Evaluate

Tabulate

Visualise

5–10 minReported time for the procurement user to build her own agent.

08 · Trust — human-in-the-loop

Fast is useful. Verified is safe.

“AgentSea handles up to Restricted, Sensitive (Normal) information. Always verify responses.”

Check the source

Can the answer point to the guideline section or uploaded evidence?

Check the output

Review facts, calculations, tone and formatting before using the deliverable.

Escalate uncertainty

If the source does not cover it, say “not covered” — never invent an answer.

09 · Guided build — transition to hands-on

Now we build the Guideline Assistant — together.

48minutes hands-on

One shared build. Synthetic data. Everyone leaves with a working pattern.

STEP 01

Configure the agent

Paste the F2 system prompt. Add role, rules and “not covered” behaviour.

STEP 02

Add the skill

Select Document Analysis so the agent can read the synthetic guideline.

STEP 03

Ground + test

Upload the guideline. Ask one covered and one deliberately uncovered question.

STEP 04

Create the briefing

Use Document Builder to return a plain-language, brand-compliant output.

Success = cited answer · “not covered” · polished briefing