SIXIVEPreparing your workspace
Sixive Workspace / Home
YOUR INTELLIGENCE WORKSPACE

วันนี้อยากให้ AI ช่วยอะไร?

เครื่องมือเดิมอยู่ครบ เลือกงานแล้วเริ่มได้จากที่นี่

AGENT FACTORY / THE ORIGINAL SIXIVE

จากแชตที่ได้ผล
สู่ Agent ที่ใช้ซ้ำได้

ดึงวิธีทำงาน เห็น Flow ตรวจข้อมูลและสิทธิ์ แล้วทดสอบด้วยโจทย์ใหม่

WORKFLOW → REUSABLE AGENT6 ขั้นตอนเดิม
  1. 01EXTRACTดึงวิธีทำงาน
  2. 02IMPORTนำ Workflow เข้า
  3. 03VERIFYตรวจความถูกต้อง
  4. 04PROTECTแยกข้อมูลส่วนตัว
  5. 05TESTลองด้วยข้อมูลใหม่
  6. 06ACTIVATEเก็บไว้ใช้ซ้ำ

Workflow เป็นส่วนหนึ่งของ Agent · ต้องตรวจและทดสอบก่อนใช้งานจริง

เลือกทางเริ่มต้น

เรียนรู้ → เตรียมงาน → ทดลอง → ใช้ซ้ำ

งานของคุณ

0Agents
0Skills
0Published locally
0Runs
ยังไม่มี Agent ที่บันทึก เริ่มจาก Agent Factory ได้เลย

สำรวจและตั้งค่า

Codex · Claude Code · n8n · OpenClaw · Hermes →
SIXIVE · INTELLIGENCE THAT WORKS

Extract how
you work.

Turn a real AI conversation into a verified, reusable Agent — keep the decision logic that makes it valuable, while separating company secrets before anything becomes public.

Core promise: Extract how you work. Turn real work into a working Agent. · Private by default.
Sixive ScoreSix example dimensions
Run economicsShown from your selected model
Gemini BYOKBring your own provider key
● Private by defaultExample flow

Market Research Agent

Example workflow: gather sources, compare findings, and ask for human review before delivery.

1Collect current sources
2Filter by business impact
3Compare against baseline
4Human approval → final output
Sixive intelligence core logo
Sixive Brand Core

Intelligence
that works.

Sixive is an Intelligence Operating System for business: an intelligence layer that turns real work, decisions and company knowledge into trusted Agents that people can verify, control and improve.

MissionTurn operational intelligence into trusted, executable action.
VisionEvery business can turn its best ways of working into AI Agents that improve with use.
01 UnderstandContext
02 DecideJudgment
03 ExecuteAction
04 MeasureOutcome
05 RememberMemory
06 ImproveLearning
Six core values
Evidence before assumptionUnknown is better than invented certainty.
Private by defaultPublic only by deliberate, reviewed action.
Humans stay in controlApproval and agency remain visible.
Outcomes over outputMeasure successful work, not token volume.
Connect, don’t lock inModels, tools and data should remain portable.
Improve every runUse feedback and telemetry to get better.
App icon, bookmark icon and home-screen identity use the same Sixive mark.
Living interface

Watch the work
before you buy it.

Browse workflow examples and open the runner. Demo mode shows a sample; Gemini BYOK calls your provider when that example supports it.

Agent · Competitive Intelligence● GOOGLE SEARCH

Competitor Intelligence Agent

Finds current competitors, builds an evidence-backed comparison table, verifies unsupported claims and prepares a Google Sheets-ready handoff.

Discover→Research→Compare→Verify→Report
Try liveBYOK Search grounding
Agent · Research● DEMO OR BYOK

Financial Research Agent

Turns a company or ticker into a decision-ready brief through research, comparison, evaluation and reporting.

Research→Compare→Evaluate→Report
$0.11Example median run
Team · Content● APPROVAL

Content Growth Team

Three coordinated roles pass context forward and stop for human approval before the final editorial pass.

Researcher→Writer→Approve→Editor
3 rolesCoordinated runtime
Office · Marketing● MULTI-ROLE

E-commerce Marketing Office

A department-scale preview that hands work across lead, research, content and analytics roles.

Lead→Research→Content→Analytics
6 AgentsOffice concept
Agent · Sales● DEMO OR BYOK

Outbound Sales Agent

Understands lead context, selects an outreach angle, drafts a message and prepares a follow-up path.

Context→Angle→Draft→Follow-up
Pay / runMarketplace mode
Office · Research● EVIDENCE-FIRST

Research Desk Office

Scout, specialist analysis, verification and executive briefing presented as one coordinated research cycle.

Scout→Analysts→Verifier→Brief
5 AgentsResearch office
The product

Learn the tools. See the flow. Reuse what works.

SIXIVE is a shared learning library and workspace: understand AI tools, make a workflow visible, test it with new input, and keep useful Agents to run again.

01 · EXTRACT

Find the real workflow.

Use the Sixive extraction prompt inside a ChatGPT, Claude, Gemini or other conversation where real work was already completed.

Evidence-first
02 · IMPORT

Bring back Workflow DNA.

Import process, decisions, corrections, tools, loops and approved outputs — not the entire raw conversation.

Structured DNA
03 · VERIFY

See why every step exists.

Each node points back to evidence. Unknown stays unknown. Suggested improvements remain separate.

No invented steps
04 · PROTECT

Keep logic. Separate secrets.

Keep reusable logic, abstract private parameters, keep company context private and block credentials.

Company-safe
05 · TEST

Run a new job.

Watch the real execution trace. Compare model quality, reliability, speed and cost before activation.

Live runtime
06 · ACTIVATE

Use it again. Share it when ready.

Keep it private and reuse it. Public publishing, paid rentals and creator payouts are future features; they are not enabled in this beta.

Private / Public
Step 03 · Verify

Workflow Graph

The graph is evidence-backed. Public logic stays visible; private parameters remain isolated.

TriggerNew research request
→
ResearchObserved step · evidence attached
→
Decision[COMPANY_POLICY] parameterized
→
DraftOutput format verified
→
ApproveHuman checkpoint
5 nodes · 4 evidenced · 1 human checkpoint
Step 05 · Test

Live Execution Trace

During a real run these states will be driven by runtime events — not decorative animation.

ResearchWaiting
→
AnalyzeWaiting
→
WriteWaiting
→
ReviewWaiting
Illustrative overview · the Agent Factory below compiles Workflow DNA into the live runner.
Privacy & confidentiality

Private by default.
Public by deliberate action.

Sixive should not make an Agent generic by deleting everything useful. Instead, it separates reusable decision logic from company-specific context.

Publish Firewall

Keep the intelligence. Move the secrets.

KEEP

Reusable process and decision logic that can safely travel with the Agent.

ABSTRACT

Convert company-specific values into configurable parameters.

PRIVATE

Internal context remains available only to its authorized owner/workspace.

BLOCK

Credentials, keys and clearly disallowed secrets never enter a public Agent.

One extractor. Two privacy views.
Both modes reconstruct the same workflow, evidence, decisions, corrections, loops, acceptance criteria and runtime requirements. Only the confidentiality policy differs.
Recommended: preserves decision logic while replacing confidential company values with placeholders before import or publishing.
Safe for the normal Sixive import review flow.
Sixive Agent Benchmark · demo table

An Agent should prove itself
on more than one model.

Creators can turn providers on or off. Sixive benchmarks the Agent on six practical dimensions and shows the run economics next to the result.

Sixive Score · Example

Research Agent v2.4

Outcome Quality94
Reliability98
Cost Efficiency91
Speed86
Agentic Execution96
User Success93

No single “best Agent” score. Sixive keeps the six dimensions visible so people can choose quality, cost, speed or operational reliability for the job they actually have.

Model Lab

Turn providers on/off for this Agent benchmark.

6 enabled
ProviderQualitySuccessMedian runCost / success
OpenAI9699%28s$0.18
Grok9396%25s$0.14
Claude9799%34s$0.23
Qwen BEST VALUE9398%24s$0.07
MiniMax9196%20s$0.06
DeepSeek9497%29s$0.08

Demo values for interface design only. Production scores should come from reproducible Sixive tests + real verified runs.

Agent gallery · marketplace direction

Explore reusable work.
See the flow behind it.

Find examples, understand each workflow, and try the ones marked for demo or BYOK. Payments, public rentals and creator payouts are not available in this beta.

▶
Explore examples before deciding whether an Agent fits.

Check each Agent's status: Demo shows a sample; BYOK can call the connected provider. External actions need their own connector.

1 · Pick an example2 · Demo or BYOK3 · Watch the graph4 · Review the result
Illustrative gallery: creator names, run counts, ratings, benchmark scores, costs and prices on these sample cards are demonstration data, not verified customer results or live offers. Purchase options below describe future models; checkout and payouts are disabled.
6 examples · sample cards, not live listings
01 · Buy once · future

Illustrative option for a protected Agent license; runtime would remain separate.

02 · Subscribe · future

Illustrative option for ongoing improvements to an Agent, Team or Office.

03 · Pay per run · future

Illustrative option to pay for a capability without buying its underlying Agent DNA.

AGENTBUY ONCE

Financial Research Agent

by @mtk · example creator

Turns a company or ticker into a source-backed research brief with comparison, risks and a human-readable conclusion.

Research→Compare→Evaluate→Report
18.5Kruns
97.8%success
$0.11median / run
31 secruntime
SIXIVE SCORE6 dimensions
94Q
98R
91E
86S
96A
93U
Best value modelQwen
Highest qualityClaude
$39license · runtime separate
Protected
AGENTPAY / RUN

Outbound Sales Agent

by @nina · example creator

Researches a lead, selects an angle, drafts outreach, and adapts follow-up based on the response.

Lead→Research→Draft→Follow-up
7.2Kruns
96.1%success
$0.07median / run
18 secruntime
SIXIVE SCORE6 dimensions
91Q
96R
96E
94S
92A
90U
RecommendedDeepSeek
Also testedOpenAI · Qwen
$0.29per successful run + compute included
Try first
TEAMSUBSCRIPTION

Content Growth Team

by @studiox · example creator

Three coordinated Agents: Researcher → Writer → Editor, with an approval checkpoint before final delivery.

Researcher→Writer→Editor
4.8Kteam runs
95.4%success
$0.32median / job
74 secruntime
SIXIVE SCOREteam benchmark
95Q
95R
88E
79S
94A
95U
Model mixQwen + Claude
Human checkpointBefore publish
$19/ month · runtime separate
3 Agents
TEAMBUY ONCE

Product Launch Team

by @launchlab · example creator

Market research, positioning, launch copy and campaign QA packaged as one reusable coordinated workflow.

Research→Strategy→Copy→QA
1.9Kteam runs
94.6%success
$0.68median / job
2m 08sruntime
SIXIVE SCOREteam benchmark
96Q
94R
84E
73S
93A
94U
Recommended mixOpenAI + Qwen
Private parametersBrand + product data
$129team license · runtime separate
Protected
OFFICESUBSCRIPTION

E-commerce Marketing Office

by @growthops · ✓ Verified Studio

A ready-to-run department with Lead, Research, Content, Social, Analytics and approval workflows.

Lead→Research→Content→Analytics
6 Agentsincluded
14workflows
$2.41median / cycle
11mfull cycle
SIXIVE SCOREoffice benchmark
93Q
97R
86E
78S
97A
92U
RouterSixive Auto · 4 models
DestinationHire into Sixive OS
$99/ month · compute separate
Office ready
OFFICEPAY / CYCLE

Research Desk Office

by @alpha · ResearchDesk beta

Continuous research desk that collects sources, assigns specialist Agents, verifies findings and delivers an executive brief.

Scout→Analysts→Verifier→Brief
5 Agentsincluded
Betaruntime
BYOKmodel cost
Livetrace
EVIDENCE GATEruntime audit
✓Claims
✓Sources
✓Math
✓Assume
✓Verify
↻Revise
Live researchGemini + Google Search
VerifierCan reject claims
$4.90per completed research cycle
Protected
No listings in this filter yet.
Creator marketplace · future direction

Show what your Agent
actually did.

The goal is to help creators show evidence from real runs and earn when others use their Agents. Marketplace payments and verified creator revenue are not live in this beta.

The creator profile and all figures below are fictional UI examples; SIXIVE has no verified creator earnings or marketplace sales yet.

M

@mtk · example profile

Research · Strategy · Content Systems
Profile concept

38.4KAgent runs
12Published Agents
71%Repeat users
$8.4KExample revenue
Example Agent

Financial Research Agent

Illustrative profile metrics · not verified results
$39Agent license
$0.11Median cost / run
31 secMedian runtime
97.8%Successful runs
72%Repeat usage
Recommended modelQwen · Best Value
Highest quality testedClaude
Last benchmark2 days ago
Privacy status✓ Publish Firewall passed
ReviewsOnly after verified purchase/run
Sixive Founder Access · Mint

Own your Sixive identity.
Unlock the 6,666 Prompt Library.

The Founder Pass borrows the strongest logic from the original 6PACK mint: a live pricing curve, a collectible AI spec card and wallet identity — now tied to useful Sixive access rather than a decorative collectible alone.

SIXIVE · AI FOUNDER PASS
#0001BASE · SIXIVE
AGENT
FOUNDER
Verified intelligence identity
94Quality
98Reliability
91Cost Efficiency
86Speed
96Agentic
93User Success
STATUS
FOUNDER
LIBRARY
6,666
ACCESS
CREATOR+
SIXIVE INTELLIGENCE NETWORK
PRIVATE BY
DEFAULT.
Public by deliberate action.
Founder identity · Marketplace access
Prompt Library entitlement · Creator badge
Agent benchmark profile
Hover the card to pause · Six Pack specs map to the Sixive Agent Benchmark
#1#300#700#1000#1500
DEMO MODE · PAYMENT RAILS NOT CONNECTED

Founder Pass + 6,666 Business Prompts

One access product, two payment rails: crypto through Base when the production contract is configured, or card checkout when a payment processor is connected.

6,666 Prompt LibraryBusiness recipes organized by job, model fit, expected output and future “Turn into Agent” action.
Founder Creator IdentityFounder badge on profile and Marketplace identity when the entitlement system is connected.
AI Six Pack CardSix benchmark dimensions presented like a computer specification card — driven by Agent data in production.
Marketplace AccessCreate, benchmark, publish and later install Agents, Teams and Offices into Sixive OS.
Current curve price
0.000100 ETHDemo USD estimate
Minted
0
0.0001 ETH base+0.5% / mint
Pay with crypto · jurisdiction-gated

Demo only. Production must use an approved legal/payment structure for the target jurisdiction, read the quote on-chain before signing, and verify the entitlement server-side.

1
0.000100 ETHCurrent demo total
Production requirement: deployed contract for crypto, verified payment webhook for cards, entitlement database, terms and jurisdiction-specific legal review. This build simulates the mint locally.
6,666

Prompt Library should be a searchable product — not a PDF dump. Each recipe can carry use case, input, expected output, recommended model, cost estimate, example and “Turn into Agent”.

Sixive Agent Factory · v0.17 · Discovery + Research

Conversation → Workflow DNA → Working Agent.

This local functional build closes the missing creator loop: import Workflow DNA, verify and edit the graph, protect company-specific context, compile it into the live runner, then save privately or publish a sanitized Marketplace package.

0My Agents
0Published
0Recorded runs
Functional boundary

What is real in this file?

Real: JSON import/validation, editable graph, client-side confidentiality classification, runtime compilation, Demo/BYOK execution, human approvals, ResearchDesk competitor search, email sign-in, profiles, RLS-protected Agent/Run cloud sync, feedback, analytics and local fallback.

Still production-gated: encrypted BYOK vault, payments, server-side confidentiality service, unattended Gmail actions, and Google Sheets/Drive/Docs/Calendar runtime adapters.

No saved Agents yet. Build your first one from Workflow DNA.
Pricing · pilot proposal

Learn free. Pay when
your workflows save real work.

SIXIVE is free to explore in this beta. A ฿199/month Creator plan is a price to test after saved-agent reuse and creator tools prove useful. No subscription or marketplace payment is active today.

Explore

฿0 during beta
  • AI Library and workflow examples
  • Prepare tasks for Codex / Claude Code
  • Build and test with your own model key
  • Model/API usage may cost extra

Creator · planned

฿199 / month · test price
  • Only after recurring value is proven
  • Saved Agents and repeat runs
  • Run history and outcome review
  • Potential creator portfolio and sales tools

Hosted runs · later

Usage shown before run
  • Separate model and tool costs
  • Hosting fee shown before execution
  • ฿399/month only to test once useful
  • Public Agent rental needs creator payouts

No real money is collected here. A subscription should pay for ongoing workspace value; AI, connector and hosted-runtime costs stay visible and separate. Creator payouts require working payments, metering and refunds first.

The other half of Sixive

Create here.
Run the company in Sixive OS.

Creators build and prove Agents here. Companies can later hire those Agents into Sixive Office / Full OS — where shared memory, departments, permissions, operations and enterprise automation live.

Sixive is readyCreate, test or explore a live Agent.
Gemini BYOK · pilot

Run Sixive with your Gemini key

Google Gemini Developer APIDirect generateContent call · x-goog-api-key authentication
ONLY PROVIDER ENABLED
Advanced settings Token limit · connection · cost estimate
Preset rates are for local estimation only. Gemini 3.8 Flash is $0.75 input / $3.75 output per 1M tokens through Dec 31, 2026 on the paid Standard tier. Thinking tokens are billed as output. Check current Google pricing ↗
Security note: this local test sends your key directly from this browser to Google. The key is not intentionally persisted by Sixive and is cleared on refresh. Google recommends that production web apps do not expose API keys client-side; production Sixive should proxy requests through a secured backend / secret vault.
Privacy: Google may use unpaid API inputs and responses to improve its products. Do not send personal, confidential or sensitive information to unpaid services. Check the billing status of your Google project. Provider data terms ↗
Current key requirement: use a current Google AI Studio Gemini auth key. Older standard keys can be rejected under Google’s 2026 key migration rules.
Not tested yet.
Marketplace checkout

Asset

Choose a payment rail. This single-file build demonstrates entitlement UX; it does not charge a card or transfer crypto.

CardProduction: verified payment webhook.
Crypto · BaseJurisdiction-gated. Production requires legal/provider configuration.
Sixive CreditsUse runtime balance / prepaid credits.
Before activationAsset price and runtime cost are separate. Model/API/tool usage may create additional charges unless the listing explicitly includes them.
Marketplace Listing

Purchase model
Observed / estimated run cost
Median runtime
Recommended model / mix
Verified runtime volume
Marketplace privacy✓ Private source context not included
Review eligibilityVerified purchase or successful run
Try before buySandbox enabled