Private preview is open to a few more founding customers.Apply
Context · The team-context layer

Context is the difference between AI that writes and AI that writes like you.

Sempleo treats context as infrastructure — curated, versioned, reviewed, and attached automatically to every agent run. Five layers. One coherent system. No prompt gymnastics.

Why context

A prompt is the smallest possible
piece of context.

Every enterprise has tried the long prompt. It tops out at a few thousand words, goes stale the moment something changes, and still can’t tell the model who the user is or which client they’re writing to. The fix isn’t more prompt. It’s a layered context model that assembles the right facts, from the right sources, for the right task — and cites every one.

That’s what the rest of this page describes.

The five layers

Layered by scope.
Governed by the people who know.

01
Scope · Organization-wide

Company

Mission, brand voice, policies, compliance posture. The things every agent should know before saying a word.

Owned byCentral ops · Brand · Legal · Compliance
company.brand-voiceVerified · Locked
Tone
Direct, plain, evidence-led. Never hype.
Person
First-person plural (“we”) externally; second-person internally.
Forbidden
“unlock”“leverage”“revolutionize”“game-changing”
Signoffs
“Best,” (external) · “—” (internal)
Em dash
Yes, unspaced. Oxford comma: yes.
company.policiesLegal-owned
# Escalation policy · excerpt
trigger: client mentions litigation, regulator,
         breach, injury, data loss
action: route to Legal review
         within: 2h
         notify: Legal lead, Account lead
# Agents must surface this — never draft a reply directly.
EntrySourceQualityUpdated
Brand voice & style guide
company.brand-voice
Brand team · Notion
Canonical
Escalation & compliance matrix
company.policies.escalation
Legal · Notion
Canonical
Product roadmap · public
company.roadmap
Product · Jira
Standard
Org chart & responsibilities
company.org
People · Notion
Standard
02
Scope · Per team or department

Team

How your team operates. Cadences, rituals, artefacts, vocabulary, shared standards.

Owned byTeam lead · Ops partner · Anyone on the team can propose
team.marketing-ops · designed exampleVerified
# team.marketing-ops · vocabulary
our MQLs → leads who booked a demo, not form-fills
our “brief” → a one-pager + stakeholder list + acceptance criteria
our sprint → two weeks, Wed → Tue, retro on the Tue
our review → author + one peer + one outsider; written, not meeting
signoff→ “—[Name], Marketing Ops”
# Agents map generic terms to these before writing anything.
team.ritualsVerified
Sprint
2 weeks · Wed → Tue · retro Tuesday PM
Standup
Async in Slack #mkt-ops-standup · before 10:00
Weekly review
Friday 15:00 · 30 min · minutes in Notion
Planning
First day of sprint · 60 min
team.comms-styleNeeds review
Default
Written, async, linkable
Slack DMs
Only for people-stuff · everything else #public
Meetings
Only if a decision can’t be made in writing
Docs
BLUF first. TL;DR last. No walls of text.
03
Scope · Per external counterparty

Client

Everything known about an external counterparty — history, preferences, contracts, people, tone.

Owned byAccount lead · CRM sync · Engagement team contributors
client.{counterparty} · designed exampleIllustrative · gaps shown
Account lead
{account lead} · VP Marketing Ops
Engagement
3 active projects · last touch 2d ago
Tone
Formal, cautious, compliance-first. No superlatives.
Red lines
No personal data in promptsEU region onlyLegal review for any public statement
Key people
{account lead} · {legal contact} · {procurement}
Commercial
MSA signed · SOW per project · NET-45
Gaps
⌀ Missing: current OKRs · Q2 roadmap · M&A posture
EntrySourceQualityUpdated
Prior engagement notes
client.{x}.history
HubSpot · call notes
Canonical
Contract & commercial terms
client.{x}.contract
Drive · signed PDFs
Canonical
Voice & tone sample
client.{x}.tone
Gmail · approved threads
Supplemental
Stakeholder map
client.{x}.people
Account lead · manual
Thin
04
Scope · Per body of work

Project

Goals, decisions, open questions, stakeholders. The story-so-far behind any piece of work.

Owned byProject lead · Anyone on the project
project.{slug} · designed exampleIllustrative
# project.brief · excerpt
goal : Refresh the {counterparty} SMB brand for 2025 launch
success : 30% lift in unaided awareness · Q3 survey
constraints: Legal sign-off on every public asset
             EU data residency · no US-only tools
team : {lead} · 3 designers · 1 copy

# Decisions log
D-04: drop serif subhead · agreed in review
D-05: no AI imagery in launch comms · legal hold
D-06: Swedish first, English second · per client
D-07: soft launch Apr 14 · full launch Apr 28
project.open-questions3 open
Q-11
Do we include the retail wordmark in launch assets? → owner: project lead · needs client call
Q-12
Final typography: Tiempos or NHU Display? → designer review Friday
Q-13
Approval chain for the press pack? → legal to confirm
project.milestones4 / 6 done
Mar 01
Brief locked ✓
Mar 22
Concept review ✓
Apr 05
Legal pre-read ✓
Apr 12
Client review ✓
Apr 14
Soft launch · this week
Apr 28
Full launch
05
Scope · Per person

User

Role, responsibilities, writing style, signature preferences, communication tone, recurring tasks.

Owned byThe person themselves · reviewed by team lead
user.{operator} · designed exampleIllustrative
Role
Senior Brand Director · Marketing · Stockholm
Reports to
VP Brand
Voice
Concise, direct, evidence-led. Rhetorical questions off.
Default tone
Warm-professional externally · dry-direct internally
Signoff
“Best, {name}” (external) · “—{initial}” (internal)
Prefers
Short emails · bulleted when there’s a list · BLUF first
Avoid
Emojis in client comms · exclamation points · “just” & “super”
user.{operator} · writing samplesIllustrative
# sample.sent · to counterparty
Hi {counterparty},

Quick update on the launch assets.

- Press pack: drafted, with legal Thursday.
- Film: two cuts, 30s and 60s, previewing Fri.
- Socials: paused until legal clears the pack.

Flagging one thing: the retail wordmark in frame 12
— can we confirm we’re keeping it? Happy to drop if not.

Best,
{name}
How it runs

Every run assembles
only the layers it needs.

Request

“Draft a reply to the counterparty.”

A user asks an agent to do something account-owned. The agent knows which layers are in scope.

  • agent: Email Drafter
  • user: {operator}
  • thread: launch
Context picked

4 of 5 layers active

Company, team, client, and user contexts attach. Project context skipped — this thread isn’t tied to one.

  • ✓ company · voice + policy
  • ✓ team · marketing-ops
  • ✓ client · {counterparty} (subset)
  • — project · not in scope
  • ✓ user · {operator}
Output

Drafted + cited

Every claim links to its source in the context. The draft sits in the operator’s inbox, not in the thread.

  • Reply drafted
  • Citations · 4 layers
  • Voice match gate
  • Quality gate: ✓ passed
Designed illustration. Real numbers depend on model and source size.

Citations, not opinions. Every agent output comes with the layers and entries that shaped it. Open a draft and you can see exactly which sentence came from which source — and when it was last verified.

Quality

Authority plus freshness.
Agents read both before they write.

Canonical

The team’s source of truth

Brand voice, escalation policy, locked commercial terms. Written by the layer’s owner and signed off; agents weight canonical entries above everything else.

Signed off by layer owner
Standard

Day-to-day working context

Team rituals, project constraints, account history, user preferences. The bulk of every workspace. Reviewed on the layer’s own cadence, not every week.

Team-owned, reviewable
Supplemental

Evidence, not verdict

Agent-proposed entries and raw samples — sent emails, call transcripts, wiki chunks. Agents cite them, but canonical and standard entries override them.

Proposed · promote or discard
Freshness

Every field has a last-updated

Each field shows an Empty / Thin / Sharp quality dot and a freshness timestamp. When a field goes stale, agents surface the gap in the run instead of guessing from old data.

Per-field · visible in every run
Context health

A dashboard,
not a log.

workspace · context healthLive
Company
100%
4 entries · all verified
Team
84%
18 entries · 3 need review
Client
62%
74 entries · 12 thin fields
Project
91%
32 entries · 1 stale

Rollup per layer. Drill into a layer to see the field list sorted by health, or open a specific entry to see the per-field quality dot and its freshness stamp.

client.{counterparty}.tone · assistThin
Quality
Thin · 42% scored by Haiku
Freshness
Last updated 42 days ago
Authority
Supplemental
Owner
Account lead
Assist · proposed fillFormal, cautious, compliance-first. Prefers written decisions over calls. Avoids superlatives; flags anything that needs Legal. Drawn from 18 approved sent threads over the last quarter.
ApproveEditReject

Thin fields get assisted, not guessed. Every field has a Haiku-scored quality dot and a freshness stamp. Where a field is thin, “Assist” proposes a fill from your existing artefacts — a human approves, edits, or rejects. The per-agent readiness table shows which agents are context-ready and which are blocked on which fields.

How we think

The opinions behind
the architecture.

PR-01

Context is infrastructure, not a prompt.

It has a schema, a review process, an audit trail, and a health metric. Treat it like a production database — because it is one.

In practice: every entry has an owner, an authority level, and a freshness stamp.
PR-02

The best context is curated, not crawled.

A scraped Notion page will never beat three sentences written by the account lead. We make human curation cheap, not optional.

In practice: agents draft context entries too — humans approve them.
PR-03

Layers scope who owns what.

Brand doesn’t own the client record; the account lead doesn’t own the brand voice. The layer model matches how teams already work.

In practice: five owners per workspace, five review cadences.
PR-04

Cite every claim, every time.

No unsourced sentence ships. If a model asserts something, the entry that produced it is one click away. Auditability is the floor, not a feature.

In practice: citations are a first-class field on every output.
PR-05

Staleness is a failure mode.

Out-of-date context is worse than no context. Every entry has a freshness window; past it, agents surface a gap rather than guess.

In practice: context health is a first-class dashboard, not a log file.
PR-06

Context belongs to the team.

Not to the vendor, not to the model, not to the individual. Per-tenant storage, per-tenant embeddings, no training on your data — ever.

In practice: data of record stays in your systems; Sempleo references it.
FAQ

The questions teams ask
in the first conversation.

FAQ-01
How is this different from RAG?
RAG is a retrieval mechanism; we use it under the hood. Context is the governance model that sits on top — who owns what, what quality state each entry is in, how often it’s reviewed, which layers attach to which task. RAG answers “find me something like this”; context answers “what should this agent actually know before it writes a word?”
FAQ-02
Do we have to curate everything manually?
No. Agents propose context entries from your existing artefacts — sent emails, Notion pages, CRM records, call transcripts. A human reviews before it gets the verified stamp. The goal is to make the human review cheap, not to eliminate it.
FAQ-03
What happens when context goes stale?
Every field carries a last-updated stamp and an Empty / Thin / Sharp quality dot. When an agent runs, it sees both — and if a field is past the freshness window its layer owner set, the agent surfaces the gap in the run rather than guessing from old data.
FAQ-04
Who can see what?
Layers carry access control. Company context is visible to the whole workspace; team context to the team; client context to the account team; project context to the project; user context to the user and their manager. No workaround — if an agent doesn’t have visibility, neither does its output.
FAQ-05
Does it work with the models we already use?
The context layer is model-agnostic. Run on Sempleo’s managed default, or route through your own Anthropic or OpenAI account on Enterprise. The assembled context gets serialized into whatever shape your model expects; switching providers doesn’t mean re-curating context.
FAQ-06
What’s the ramp-up like?
Kickoff to production in four moves: connect sources on day one; seed company and team layers in week one; install the first agents from the starter catalogue in weeks two and three; go live in week four. The founder sits with you for the whole first pass. Context grows with use after that — every reviewed output becomes a voice sample.
Everyone’s racing to the next model. I’m betting the bottleneck has already moved — it’s not capability, it’s the shape of what you hand the model. Company voice, team rituals, client history, project constraints. Five layers that are almost never written down anywhere an agent can read. If Sempleo is right about anything, it’s that curating those layers is the real work — and the model you use is an implementation detail.
Lasse Nørby
Lasse NørbyFounder · Sempleo
Sempleo

Shape the team-context
layer with us.

We're onboarding a small cohort of founding customers to deploy Sempleo on real workflows. A 45-minute call with the founder — you leave with a plan; we leave with the shape of how your team actually works.