01
Do and verify the work
Atlas gathers evidence, completes the request, and checks the real result before treating the approach as proven.
01 · Purpose
What Atlas is
Atlas is a deeply informed digital teammate: an agent that retains rich operational context and can use it to help the team understand what is happening, decide what to do next, and move work forward.
It connects knowledge across approved shared sources, keeps that understanding current, and turns context into practical action. Atlas can answer questions, generate reports, analyse patterns, execute tasks across platforms, build automations and workflows, and create or run reusable skills.
Atlas is not limited to the Creative lane as a product vision. Omaze DE Creative is its current proving ground, while the longer-term goal is to support people and workflows across the wider team.
Context that can act
Atlas combines persistent team knowledge with execution. It should ground its work in evidence, show what it used, verify consequential actions, and continuously improve how it works.
02 · Self-improvement
Teach Atlas once. Reuse and automate the workflow.
Atlas is powered by the Hermes self-improvement loop. It can turn a proven process, or a correction from the team, into a reusable skill rather than relearning the same workflow in every conversation.
02
Capture the reusable method
It records the procedure, required sources, permissions, approval gates, verification steps and known pitfalls in an Atlas skill.
03
Learn from team interaction
If Atlas gets something wrong, a teammate can correct it. If someone has useful context that Atlas does not yet know, they can share it directly in the conversation. Atlas can store durable, relevant context so it is available to the team in future interactions.
04
Reuse and improve it
Future sessions can use the updated context and load the improved skill. Atlas preserves the useful correction or new information instead of forcing the team to explain it again.
05
Automate only when approved
A skill can be run on demand or connected to a scheduled task or event-driven automation. Creating the skill does not automatically schedule it or grant permission to write to external systems.
Memory
Stable facts, preferences, access details and standing boundaries.
Context vault
Shared operational knowledge, decisions, source maps, campaign context and evidence. Authorised teammates can add or correct durable context through normal conversation.
Skills
Reusable execution playbooks describing how Atlas performs a specific workflow safely and consistently.
Automations
Approved skills connected to schedules or events so useful workflows can run reliably for the team.
Teach Atlas through conversation
You do not need to edit a database or write a special document. Tell Atlas what it got wrong, explain a missing piece of context, or give it a source to ingest. Atlas can preserve durable information for future use while distinguishing a teammate’s contribution from independently verified source evidence.
You can create skills with Atlas
Describe the workflow in normal language. Atlas can inspect related Cowork skills, n8n workflows and existing Atlas capabilities, propose a safe design, build it, dry-run it, and separately ask before enabling automation or consequential writes.
Custom Exactius skills already available
These are team-specific capabilities built around real Exactius workflows, not generic chatbot features.
Creative operations ingestion
ActivePulls Creative activity from approved operational sources, updates house hubs, links evidence, detects action candidates and produces status updates.
Purpose: Keep the Omaze DE Creative operating picture current without manually assembling every source.
Omaze learnings ingestion
ActiveTurns weekly decks and meeting transcripts into structured learnings, a learning agenda and a testable roadmap with evidence requirements.
Purpose: Preserve what the team is learning, not just what happened.
CRM brief write-back
AvailableCopies approved final copy from a working Google Doc into the original CRM strategy brief while preserving its tabs, tables and wireframes.
Safety: Runs a structural dry-run and requires approval before writing.
Workflow porting
AvailableReads existing n8n workflows and Cowork skills, understands their intent and dependencies, and adapts approved workflows into Atlas skills and automations.
Purpose: Reuse good work instead of rebuilding the process from scratch.
Public and secure page publishing
AvailablePublishes approved reports, dashboards and static artifacts as public links or secure Exactius pages protected by an email/domain allowlist.
Safety: Scans content and confirms the visibility choice before publishing.
03 · Current understanding
What Atlas understands today
The current context is specific to Omaze DE Creative—not a general archive of every Exactius or Omaze project.
Creative workflow
Request → triage → brief → production → internal review → partner approval → export/delivery → performance feedback.
Campaign structure
Launch, BV/bonus phases, Early Close, Close, Big Swings and ongoing/evergreen work, linked by house.
Creative outputs
Static, video, copy and concept work across Meta and secondary channels, with documented naming conventions.
Live house hubs
Current operational hubs for H5, H6 and H7 with linked tasks, assets, stages, pending approvals, deadlines and recent mentions.
Source authority
Which source answers which question—for example, ClickUp for task stage and Figma comments for static-review evidence.
Campaign memory
Stable workflow facts, source documents, meeting operations, action-item candidates and curated learnings with supporting data.
People and roles
Who team members are, their responsibilities, the workstreams they support and identity details needed to avoid mixing up similarly named people.
Meeting purpose
What each recurring meeting is for, which topics belong there, who participates and how decisions or actions should flow into the wider operation.
Team language and naming
Creative naming conventions, house and phase codes, workflow terminology and the source documents that define how work should be classified.
04 · Connected systems
Access today, with the real limits
“Connected” does not mean “everything in the system.” Atlas can only use the accounts, channels, files, projects and write permissions that are authorised and shared with it.
ClickUp
Read + writeHas normal-user access as Atlas. It can read, create, edit, assign, move and close tasks, and add comments across the authorised workspace.
Boundary: ClickUp remains authoritative. Atlas should confirm consequential or ambiguous changes before writing, then read the task back to verify the result.
Slack
Live nowScheduled ingestion reads four approved shared/project channels, applies a DE filter and surfaces operational requests with message evidence.
Boundary: Slack messages are signals, not canonical task status. DMs are excluded from shared-context ingestion.
Figma
PartialReads shared Omaze DE house-file comments with author, time and node context. H7 pages/comments are available; H6 comments are available.
Gap: H6 full-file export is blocked, and performance-file keys for earlier houses have not been supplied.
Frame.io
Live nowRead access to the Omaze DE project, including folders, files and authored comments. Asset paths can be linked to tasks and house hubs.
Boundary: Read-only. Comment language is evidence; formal approval logic is not assumed where the source does not prove it.
Google Workspace
Live nowAuthenticated access to the shared Creative source set, including the source map, H6/H7 narratives, H6 Big Swing deck and H6/H7 messaging matrices.
Verified: Shared Drive discovery, metadata and export now work correctly. A separate copy-working document still needs a canonical link, and the Gemini Creative-call notes remain unshared.
Fireflies
Live nowReads meeting transcripts and filters them against the documented Creative meeting guide to capture asks, commitments and asset mentions.
Boundary: Only available transcripts and matching meetings can be used. Atlas separates operational actions from learnings.
GitHub + context vault
Live nowA Git-backed, curated context vault holds source summaries, house hubs, action candidates, meeting operations and learnings with evidence links.
Boundary: It is a curated layer, not a raw mirror of every message, comment or transcript.
Scheduled tasks + automations
Build + runCan design, build and maintain scheduled tasks, event-driven automations and reusable workflows that operate across authorised platforms.
Default: Atlas should define the trigger, permissions, approval gates, verification and delivery clearly before enabling a workflow.
n8n
Source accessCan read Exactius n8n workflow definitions and execution history to understand existing triggers, node logic, mappings and platform dependencies.
Boundary: n8n is source material for porting. Atlas rebuilds approved workflows as scheduled tasks and automations rather than treating n8n as the default runtime.
Cowork plugins repo
Source accessCan read the private Exactius/cowork-plugins repo and use its user-created skills as source material for new Atlas capabilities and automations.
Boundary: Preserve the skill’s intent, safety rules and judgment. Port it into Atlas/Hermes rather than silently changing behavior.
Shareable pages
Public + secureCan publish approved reports, dashboards and static artifacts as either public links or secure Exactius pages protected by email one-time PIN access.
Choice required: Atlas confirms public versus secure before publishing. Secure pages use an approved email/domain allowlist and have no public origin bypass.
Source hierarchy matters. If Slack says “approved” but ClickUp still says “in partner approval,” Atlas should flag the conflict—not silently overwrite the ClickUp status.
05 · Practical use
What the Creative team can ask right now
Good requests have a house, phase, task, channel or time window. Atlas can still help when those are missing, but it will label uncertain links rather than guessing.
01
Collaborate with Atlas as a team
Multiple teammates can work with Atlas in the same shared thread: brainstorm ideas, ask follow-up questions, bring different perspectives, request analysis and turn the discussion into concrete actions. They can also correct Atlas or contribute missing context, which Atlas can preserve for future conversations so the group keeps building on a shared understanding.
02
Find assets and source documents
Ask for the ClickUp task, Figma frame, Frame.io asset, narrative or naming source for a house/phase. Atlas can return the best available links and call out missing access.
03
Check task and campaign status
Request a house or phase roll-up, open tasks, due dates or recent stage changes. The answer should be anchored to current ClickUp evidence.
04
Track approvals and exceptions
Surface assets in partner approval, candidate approval language, unresolved review exceptions and conflicting signals. Comment-derived approvals stay labelled as candidates until stronger status evidence confirms them.
05
Catch operational requests
Atlas scans approved shared Slack channels, matching Fireflies meetings, Figma comments and Frame.io comments for explicit asks or commitments. It checks for an existing ClickUp task before surfacing new work.
06
Create and maintain ClickUp work
Atlas can check for duplicates, then create or update tasks, set ownership and dates, change status, and leave comments. For consequential or ambiguous changes it should show the intended mutation first, get approval, and verify the saved task afterward.
07
Build and run scheduled tasks and automations
Describe a recurring or event-driven workflow and Atlas can turn it into a reusable skill and automation, including source access, guardrails, approval gates, verification and delivery. It can also adapt existing n8n workflows or Cowork skills.
08
Prepare briefs and status reports
Combine current source links into a concise brief, daily status, meeting pre-read, handoff or “what did I miss?” summary. Missing sources and conflicting evidence remain visible.
09
Maintain campaign memory and learnings
Preserve stable workflow context and evidence-backed learnings across weeks. Learnings follow a structured method and should not appear without supporting data.
10
Connect activity to performance context carefully
Atlas can line up what was made, when it changed, what test was intended and what performance data was reported. It can describe associations and test evidence; it cannot prove causality unless the source and test design support it.
06 · Complementary tools
Atlas and Claude on your computer
Atlas does not replace Claude Cowork or Claude Code. The useful distinction is between a persistent cloud teammate with shared context and execution capabilities, and a local Claude session helping one person with the work in front of them.
| Area | Atlas | Claude Cowork or Code on your computer |
|---|---|---|
| Role | A persistent digital teammate serving shared team workflows. | An interactive assistant working with an individual in a local session. |
| Context | Maintains rich, shared operational context across approved sources, projects and interactions. | Primarily uses the current conversation, local project, files and services available to that person. |
| Continuity | Retains team knowledge and reusable capabilities across sessions and over time. | Optimised for the current piece of work; continuity depends on that user's sessions, projects and setup. |
| Execution | Can work across authorised platforms, execute operational tasks, generate reports, analyse activity and coordinate workflows. | Can perform powerful work with the user's local tools and connectors while that person directs the session. |
| Automation | Can build and run durable scheduled or event-driven workflows for the team. | Usually interactive and user-driven unless the person separately configures automation infrastructure. |
| Skills | Uses team-owned skills with shared context, permissions, approval gates and verification. Skills can become durable workflows. | Uses personal or project-level instructions suited to that user's files, environment and immediate work. |
| How they complement | Provides the shared context and always-available execution layer for recurring team operations. | Provides a flexible personal workspace for exploration, drafting, coding and prototyping workflows that may later be brought to Atlas. |
Personal workspace meets shared teammate
Use Claude locally when you want a powerful assistant beside you for an active piece of work. Use Atlas when the work benefits from shared context, continuity, cross-platform execution or an automation that should keep running for the team.
07 · Capability map
Live now, partial and future potential
| Capability | State | What that means |
|---|---|---|
| ClickUp status and deadline checks | Live now | Reads the DE Creative Pipeline and OM Timeline; ClickUp is the final authority. |
| Shared Slack request detection | Live now | Four approved shared/project channels are monitored with DE filtering. DMs are excluded. |
| Figma review evidence | Partial | H7 pages/comments and H6 comments are readable; H6 full export and earlier performance-file links are missing. |
| Frame.io asset and comment retrieval | Live now | Omaze DE folders, files and authored comments are readable. Atlas does not write or set approval state. |
| Google narratives and source docs | Live now | Shared Drive access is fixed and verified for the Creative source map, H6/H7 narratives, H6 Big Swing deck and H6/H7 messaging matrices. The separate copy-working document still needs a canonical link, and Creative-call notes remain unshared. |
| Meeting action capture | Live now | Available Fireflies transcripts are scanned against the Creative meeting guide for operational asks and commitments. |
| House/phase operational memory | Live now | H5–H7 hubs connect tasks, assets, approvals, deadlines, mentions and evidence. |
| ClickUp task and comment management | Read + write | Atlas can create, edit, assign, move, close and comment on tasks. Consequential or ambiguous changes require confirmation and read-back verification. |
| Scheduled tasks and automation building | Live now | Atlas can build and maintain reusable skills, scheduled tasks and event-driven workflows with explicit guardrails and verification. |
| Code and dashboard delivery | Future | Atlas could write and test code, build new dashboards and reports, and deploy approved outputs as public or secure pages as this capability is developed and governed. |
| n8n workflow inspection and porting | Live now | Reads Exactius n8n workflows and execution history, then adapts approved workflows into scheduled tasks and automations. |
| Cowork skill inspection and porting | Live now | Reads the private Exactius/cowork-plugins repo and ports approved skills while preserving intent and safeguards. |
| Google Doc brief write-back | Live now | Can dry-run and, after approval, copy finalized text from a working Doc into the original CRM brief while preserving structure and wireframes. |
| Curated learnings and performance context | Live now | Uses structured learnings rules and available reported data; avoids unsupported causal claims. |
| House/phase/format performance querying | Partial | The dedicated Violet comparison query path is not connected, and DE performance-file keys are incomplete. |
| Complete copy and Big Swing context | Partial | Blocked Drive items prevent complete coverage until files are shared and broken links are corrected. |
| Connected data agent | Future | Connect Atlas to a dedicated data agent so teammates can ask operational questions in context and receive governed insights, comparisons and visualisations from the underlying data. |
| Broader team and source coverage | Future | Expand Atlas beyond the current Creative proving ground as additional teams, files, channels, data products and formal signals are deliberately connected. |
08 · Help shape Atlas
What else should Atlas help with?
Atlas grows around real team needs. If you have a repetitive task, a report you wish existed, a source that is hard to search, a workflow that keeps breaking, or an idea for how the team could work better, bring it to the R&D team. We can explore it with you and build the right capability into Atlas.
Bring R&D a need
Describe the outcome you want. You do not need to know what the technical solution should be.
Bring us a frustrating workflow
Show the R&D team the manual steps, duplicated work or recurring handoff that consumes your time.
Bring us a source
Suggest a system, document set or data product that would make Atlas’s context, analysis or execution more useful.
Bring R&D an idea
We can explore the value, access, safety and effort with you before deciding how to build it into Atlas.
You do not need to design the solution
Start with “I wish Atlas could…” or “Our team needs help with…” and bring that need to R&D. We will work with you to understand the workflow, investigate what is possible and build the strongest ideas into Atlas as new context sources, skills, reports, dashboards or automations.
09 · Start here
Example prompts to try
Copy one, then replace the house, phase or date range. Atlas should answer with links and flag missing evidence.
“Let’s use this thread to brainstorm the next campaign idea together. Keep track of everyone’s suggestions, connect them to what we already know, surface disagreements and turn the strongest direction into clear next steps.”
Team collaboration“That context is out of date. Here is the corrected information and the source it came from. Update the shared context, tell me what changed, and use the correction in future answers.”
Correct shared context“Give me the current H7 BV3 Creative status: tasks, stages, partner approvals, exceptions and the next deadlines. Link every source.”
Status roll-up“Find the source task and review links for the H7 BV2 Value Count-up video. Tell me what is confirmed versus comment-only.”
Asset retrieval“What Creative requests appeared in approved Slack channels, Fireflies, Figma or Frame.io this week that are not already covered by ClickUp?”
Request detection“Check ClickUp for duplicates, then show me the task you propose creating, including owner, due date, status and source evidence. Wait for my approval before creating it.”
Controlled task creation“Build a scheduled task that checks the Creative Pipeline every weekday, flags overdue partner approvals, and drafts an update for approval. Show me the plan and dry-run before enabling it.”
Automation building“List the relevant n8n workflows and Cowork skills for Creative operations, then recommend which one to port into Atlas first. Do not build until I approve the plan.”
Automation source discovery“Prepare a five-minute pre-read for the next Creative status meeting: changes since the last meeting, pending approvals, deadlines and blockers.”
Meeting pre-read“I was away for three days. Summarise what changed in Omaze DE Creative, grouped by house and phase, with source links and unresolved conflicts.”
Missed context“Show the available evidence connecting the latest Creative activity to reported performance. Separate observation, test result and unsupported causality.”
Performance context“Which Creative source documents are accessible for H6 and H7, and which expected files are still blocked or incorrectly linked?”
Access checkHow this guide was checked. Capability claims were reconciled against the live Creative Lane documentation, source-access matrix, current H5–H7 hubs, action-candidate ledger, scheduled workflow state and direct integration checks. Access can change; Atlas should re-check the source before giving a time-sensitive answer.