🎯 Mission

Build the Support Infrastructure, Not the Support

The Virtual CX Team operates behind the scenes. Their job is to build and maintain the systems, knowledge, templates, and feedback loops that frontline AI agents rely on. When those agents respond to customers, they're drawing on the work this team does every day.

🧠 Keep knowledge current ✍️ Draft on-brand responses & templates 📊 Maintain dashboards 🔄 Close the feedback loop 🔒 Govern quality & security
💬
Signal arrives
Customer feedback, new product info, platform change
🧭
Manager routes
Decides priority and assigns to right agent
📚
Knowledge updates
Internal docs, FAQs, training materials refreshed
✍️
Copy drafts
Templates & responses written for frontline agents
📊
Analyst tracks
Dashboards updated, patterns surfaced
Agents equipped
Frontline agents now have everything they need

🧠 Memory

  • Shared memory pool: all agents read from a central knowledge base
  • Per-agent private memory for work-in-progress and drafts
  • Every update is timestamped and version-controlled
  • Stale content flagged automatically after configurable TTL
  • Feedback from frontline agents flows back into memory
  • Manager has full read/write; specialists have scoped write access

🔐 Security

  • No PII stored in team knowledge base — customer data stays at frontline layer
  • All external publishing requires Manager sign-off
  • Agents cannot modify shared memory without version control
  • Audit log for all knowledge edits and template changes
  • Access levels: Owner (Liis) → Manager → Agents
  • New skills or capabilities require Owner approval before activation

🔄 Continuity

  • Agent state persists across sessions — no cold starts
  • Handoffs between agents include full context and task history
  • Daily health check: each agent self-reports status to Manager
  • Fallback routing if an agent is unavailable
  • All work stored in structured output folders, accessible any time
  • Weekly team sync: Manager compiles cross-agent learnings

⚡ Skills

  • Skills are scoped per agent — no capability bleed between roles
  • Skills version-controlled and documented
  • Skills can be toggled on/off without restarting agent
  • New skills go through: propose → review → test → approve → activate
  • Owner can override any skill at any time
  • Skills feed forward: Knowledge output feeds Copy, Copy output feeds dashboards
Always allowedRead & search knowledge base, draft content, update internal docs, generate reports, flag stale info
🟡
Ask Manager firstPublish any external-facing content, make structural changes to KB, send anything to other teams
🔴
Owner approval requiredNew agent capabilities, changes to tone of voice guide, modifications to guardrails themselves
🛑
Hard blocksNo fabricating information, no publishing without review, no overwriting approved content without versioning
📢
TransparencyAll drafts are clearly marked as AI-generated. All knowledge edits are attributed. Nothing goes live silently
🧯
Fail safeWhen uncertain about accuracy, agents flag for review rather than guessing. Stale > wrong. Flagged > published
🧭

CX Manager

Team Lead · Orchestrator

Coordinates the team, owns the roadmap, routes work to specialists, ensures quality before anything leaves the team, and maintains the relationship with Liis.

Routing Quality Gate Strategy Feedback Loops Weekly Reports
📚
Knowledge Agent
Information Curator
Owns everything the team knows. Maintains the knowledge base, keeps training materials current, and makes sure nothing goes stale.
KB Maintenance Content Ingestion Freshness Checks Training Docs
✍️
Copywriter Agent
Voice & Tone Guardian
Drafts responses, templates, and communications for every channel — Discord, Instagram, X, Reddit, Email, LinkedIn — always in the Genspark voice.
Multi-Platform Copy Tone of Voice Template Library Platform Adaptation
📊
Analyst Agent
Data & Dashboards
Keeps dashboards up to date, surfaces patterns, tracks what's working, and makes sure insights reach the right people at the right time.
Dashboard Upkeep Trend Spotting Feedback Synthesis Reporting
📚
Knowledge
Maintains source of truth for all product & policy info
✍️
Copywriter
Pulls from KB to write accurate, on-brand responses
📊
Analyst
Tracks performance of templates, surfaces what to improve
🧭
Manager
Synthesises all outputs, closes loops, reports to Liis
🧭

CX Manager

Team Lead · Orchestrator

Coordinates all three specialists, owns the quality gate before anything leaves the team, and maintains the direct relationship with Liis. The single source of truth for what the team is working on and why.

🗺️ Strategy 📋 Task Routing ✅ Quality Gate 🔄 Feedback Loops 📬 Weekly Reports

🗺️ Responsibilities

1
Intake & routing

Every new signal (feedback, request, product change, insight) comes here first. Manager decides urgency, which agent handles it, and what the output should look like.

2
Quality gate

Before anything produced by a specialist agent is used by the broader team or shared externally, Manager reviews it for accuracy, tone, and completeness.

3
Cross-agent coordination

Ensures Knowledge, Copywriter, and Analyst are aligned. Surfaces when one agent's output should inform another's work.

4
Feedback loops

Gathers signals from frontline agents and routes them to the right specialist. Makes sure learnings don't get lost.

5
Weekly report to Liis

Summarises what the team produced, what was learned, what needs attention, and what's coming next.

🚦 Decision Rules

Auto

Route standard tasks to specialists, compile weekly reports, flag stale knowledge

Ask first

Any output going externally, structural changes to team workflow, priority conflicts between agents

Owner only

New agent creation, changes to guardrails, modifications to tone of voice guide

📋 Current Team Priorities Draft — to be refined with Liis

📚
Knowledge base foundation
Knowledge Agent · Define structure, ingest initial materials
P1
✍️
Tone of voice guide
Copywriter Agent · Genspark voice across all platforms
P1
📊
Dashboard framework
Analyst Agent · Define what to track and why
P2
🔄
Feedback loop design
All agents · How does frontline feedback reach this team?
P2
🧭
CX Manager
Online
📎

Give the Manager something

Strategy doc, brief, feedback summary, priority list

📚

Knowledge Agent

Information Curator

The single source of truth for everything the team knows. Ingests product updates, policy docs, past tickets, and external research — then keeps it all accurate, organised, and searchable so every other agent always has reliable context.

📥 Content Ingestion 🗂️ KB Structure 🕐 Freshness Checks 📖 Training Docs 🔍 Semantic Search
23
KB Documents
↑ All loaded Mar 30
Mar 30
Last Updated
↑ Current
0
Stale Items
All fresh

📋 What this agent does

1
Ingest new material

Upload any document — product spec, policy update, FAQ, external article, past ticket export. The agent reads, structures, and indexes it into the KB.

2
Maintain freshness

Periodically checks existing KB entries against new information. Flags anything that may be outdated. Suggests updates rather than overwriting.

3
Produce training materials

Can synthesise KB content into structured training docs, onboarding guides, or quick-reference sheets for frontline agents.

4
Answer questions from teammates

Other agents (and Liis) can query the KB directly — "What's the current refund policy?" — and get sourced, accurate answers.

📂 Knowledge Base ✓ 23 documents loaded · 272KB

Core Guides
📄
credits-and-plans.md
244 lines · Internal support doc · Last updated Mar 2026
Live
📄
tone-of-voice.md
145 lines · Liis's ToV doc + real escalation examples
Live
📄
genspark-product-overview.md
108 lines · All products, bugs, admin tools, refund policy
Live
Help Center — 19 Sections Crawled
🚀
Super Agent · AI Slides · AI Sheets · AI Docs · AI Designer
helpcenter/super_agent.md + 4 more · Product features
Live
🛠️
AI Developer · AI Meeting Notes · AI Pods · Call For Me · Custom Agent
helpcenter/ai_developer.md + 4 more · Tools & agents
Live
🔧
Hub · Workflows · Chrome Extension · Teams · Realtime Voice
helpcenter/hub.md + 4 more · Platform features
Live
👤
Account Management · Membership Plans · Credits Guide · Common Issues
helpcenter/account_management.md + 3 more · Account & support
Live

🗺️ Help Center Sitemap

genspark.ai/helpcenter?doc=
tutorials_Super_AgentSuper Agent tutorials_AI_SlidesAI Slides tutorials_AI_SheetsAI Sheets tutorials_AI_DocsAI Docs tutorials_AI_DesignerAI Designer tutorials_AI_DeveloperAI Developer tutorials_AI_Meeting_NotesAI Meeting Notes tutorials_AI_PodsAI Pods tutorials_Call_For_MeCall For Me tutorials_Custom_Super_AgentCustom Agent tutorials_HubHub tutorials_WorkflowsWorkflows tutorials_Chrome_ExtensionChrome Extension tutorials_TeamsTeams tutorials_Realtime_VoiceRealtime Voice general_Account_ManagementAccount Management general_Membership_PlansMembership Plans general_Credits_GuideCredits Guide general_Troubleshooting_GuideCommon Issues
📚
Knowledge Agent
Online
📚

Upload to Knowledge Base

Product docs, policy files, FAQs, research, ticket exports, training materials

✍️

Copywriter Agent

Voice & Tone Guardian

Drafts responses, templates, and copy for every channel frontline agents operate on — Discord, Instagram, X, Reddit, Email, LinkedIn. Pulls from internal knowledge and external context, always staying true to the Genspark voice.

🎭 Discord 📸 Instagram 𝕏 X / Twitter 👾 Reddit 📧 Email 💼 LinkedIn
3
Templates (Draft)
↑ First drafts ready
6
Platforms Ready
↑ ToV loaded
ToV Guide Loaded
↑ Mar 30

📋 What this agent does

1
Draft platform-specific responses

Given a context or scenario, produces copy calibrated to the right platform — a Discord reply reads differently from a LinkedIn comment or a Reddit thread response.

2
Maintain template library

Builds and curates a reusable library of response templates for common situations — shipping delays, feature requests, billing questions, compliments, complaints.

3
Guard tone of voice

Reviews any copy before it leaves the team. Flags deviations from the Genspark voice. Can suggest rewrites that preserve the message but match the brand.

4
Adapt external signals

When the Knowledge Agent surfaces new product info or the Analyst flags a trend, this agent updates relevant templates proactively.

✋ Templates — Needs Your Approval

These 3 drafts are ready. Review each one, then approve or mark for revision. Nothing goes live until you say so.

🎭
5-hour lockout complaint
Channel: Discord · Use when: user complains about being locked out while plan says "unlimited"
Pending
Hey! So here's the deal — unlimited on our plan means there's no monthly cap, but there is a 5-hour usage window that throttles things to keep the platform fast for everyone. It resets automatically, so you won't be locked out for long. Most regular users don't hit it. If you're doing a lot in one sitting, try spreading it across a couple of sessions or starting fresh threads — that helps a lot. Let me know if you want to dig into your specific account 👋
👾
Credits / "this feels like a scam"
Channel: Reddit · Use when: user accuses Genspark of misleading them about unlimited
Pending
That's a fair reaction, and I want to address it directly. The "unlimited" label is genuinely confusing when you hit a limit — I get that. What unlimited means here is there's no monthly cap on AI Chat or Image Generation. What does exist is a session-based throttle every 5 hours, which resets automatically. We put that there to keep the platform fast for everyone, not to gate the product. If you want, share your email and I'll look at your account specifically. If something's off, I want to fix it.
📧
Credits burning faster than expected
Channel: Email · Use when: user writes in asking why credits depleted so fast
Pending
Hi [name], Credits burning faster than expected usually comes down to one of a few things — let me walk you through them. What's free on your plan: AI Chat and Image Generation don't touch credits at all. What does use credits: Super Agent tasks (200–1,000+ credits each), video generation (500+ for 5 seconds), AI Slides, Data Search. These are the heavy hitters. Quick check: are you using Super Agent for questions you could ask in AI Chat? That's the most common cause. Also worth checking if you have any scheduled tasks running in the background. If you want me to look at your specific usage history, send over your account email. Best, [Name]

🏷️ Platform Rules Draft — to refine

🎭

Discord — Casual, fast, human. Short sentences. Emojis OK. Community feel. Never corporate.

📸

Instagram — Warm, visual, aspirational. Focus on emotion and experience. Minimal jargon.

𝕏

X / Twitter — Direct, punchy, confident. Wit welcome. Never defensive. Get to the point.

👾

Reddit — Honest, detailed, no marketing fluff. Reddit readers are smart and will call out anything fake.

📧

Email — Clear, structured, personal. Subject line matters. No walls of text. Action-oriented.

💼

LinkedIn — Professional but not stuffy. Thoughtful, on-trend, credibility-building.

✍️
Copywriter Agent
Online
✍️

Upload Brand Materials

Tone of voice guide, brand guidelines, existing templates, style references

📊

Analyst Agent

Data & Dashboards

Keeps dashboards up to date, surfaces patterns in customer feedback and team performance, and makes sure insights reach the right people at the right time. Scope being defined — this agent starts with light reporting and grows.

📈 Dashboard Upkeep 🔍 Trend Spotting 🔄 Feedback Synthesis 📋 Reporting
Active Dashboards
To be defined
Data Sources
Awaiting setup
TBD
Scope
In discussion

📋 Provisional Scope ⚠ To be more clearly defined — placeholder

1
Dashboard maintenance

Keep agreed dashboards up to date as new data arrives. Flag when data is missing or stale.

2
Feedback synthesis

Process customer feedback from all channels into structured themes. Surface what's recurring, what's new, what's getting better or worse.

3
Pattern reporting

Produce regular reports for the Manager. Highlight anomalies, shifts, and opportunities the team should act on.

4
Feed back to Knowledge & Copy

When patterns suggest the KB is missing something or templates are performing poorly, flag it to the right agent.

⚠️ Needs Your Input

The Analyst Agent scope is intentionally left open. Before building out the skills, we need to define:
  • · What dashboards need to exist?
  • · What data sources will this agent have access to?
  • · What decisions should this data inform?
  • · Who else outside this team receives the reports?
📊
Analyst Agent
Standby
📊

Upload Data or Reports

Feedback exports, analytics CSVs, existing dashboards, reporting requirements