Ad Content Studio
A copywriting agent that writes on-brand ads from a brief: it knows your brand guide, a compliance reviewer checks claims against your ad policy, and an attached workflow checks character limits and…
Runs on Google Ads
Home / Marketplace
Camaan Marketplace
Enterprise agents that understand your business. And that you own.
One engine. Knowledge we package on top: offers built once for a process, a team, a platform and an industry, installed into your organisation and yours to change.
23 offers today
Agents, workflows, reports and apps, ready to install.
Built once, run on your platformEach platform gets its own native build, tested against the same cases.
Browse by what you run
Start from the work, not from a blank agent.
Pick the process, the team, the platform and the industry. Every offer is described along the same four lines, so you can find what fits and compare like with like.
Find offers for that serve on in
23 offers in the marketplace
Campaign ROI Calculator
Works out what a campaign really returned: pulls spend from your ad platforms and attributed pipeline and revenue from the CRM, calculates ROI, ROAS, CAC and payback in code under your attribution…
Runs on HubSpot
Canonical Model Mapper
Maps a new source system's table onto your canonical data model: profiles the data, proposes every field's mapping with a confidence and a reason, checks required-field coverage, and publishes only…
Runs on Any platform
COBOL Migration Assistant
Migrates a COBOL program to a modern language the safe way: inventories it and its copybooks, sends constructs like CICS or ALTER to an architect, translates with your migration rules, and proves the…
Runs on IBM i
Control Test & Evidence Pack
Tests a control the way an auditor would: a repeatable risk-based sample, evidence collected item by item, each item judged against the control objective, and a workpaper written - with every…
Runs on Jira
Diet Management Agent
A diet coach patients can talk to: it knows their targets, conditions, allergies and preferences, reviews their food log, and builds weekly meal plans through an attached workflow that checks…
Runs on FHIR EHR
Due Diligence Questionnaire Risk Analyzer
Reads a counterparty's completed due diligence questionnaire, checks the answers against your risk policy while screening for adverse information, scores the risk with cited drivers, and routes it -…
Runs on Any platform
Failed Provisioning Remediator
Picks up a failed provisioning job, finds the root cause from logs and live state, and fixes it - automatically when the fix is on your safe list and low risk, otherwise after an operator approves…
Runs on ServiceNow
Google Ads Campaign Manager
Optimises Google Ads every day: finds wasted search terms, campaigns over target CPA and budget-limited winners with rules in code, proposes bid, budget and negative-keyword changes with reasons…
Runs on Google Ads
Incident Triage & SLA Breach Watch
Classifies every new incident, finds duplicates and the change that probably caused it, pages the on-call lead for P1s with a timeout escalation, and assigns the rest.
Runs on ServiceNow
Multilingual Service Line
A voice agent for your service line that answers in the caller's language, finds their account, checks known outages, logs requests, and - through an attached wrap-up workflow - books a human…
Runs on Zendesk
Onboarding Document Checker
Checks a new customer's onboarding file end to end: extracts the details from the documents, verifies identity and screens sanctions alongside a policy check, and routes the file - approve, ask the…
Runs on Salesforce
Order-to-Cash Revenue Leakage Monitor
Finds revenue that leaks between order and cash: matches every delivery to its order and invoice in code (not by guesswork), flags unbilled deliveries, under-billing and unexplained credits, explains…
Runs on NetSuiteSAP, in validation
Policy Assistant with Escalation
An internal policy assistant that answers only from your policies, cites them, and knows when it cannot answer: a specialist checks the exact wording, and an attached workflow drafts what it can and…
Runs on ServiceNow
Prescription Safety Checker
Checks every new e-prescription before a pharmacist verifies it: allergies, interactions, duplicate therapy and dose limits against the patient's chart and your formulary.
Runs on FHIR EHR
Process Deviation Responder
Responds to a plant process deviation the way a good shift engineer would: confirms it against the limits, reads the trend, asset state and your SOP, recommends a correction, applies small ones…
Runs on Any platform
Product Video Studio
A video producer agent that turns a short brief, and optionally a product photo, into a finished narrated video: it writes the script, voices every scene with the platform's text-to-speech, builds 3D…
Runs on Any platform
Regulatory Return Drafter
Drafts a periodic regulatory return from the ledger and the regulator's own instructions, checks every line against last period and the validation rules, and files it as a draft only after the…
Runs on Snowflake
Sales Forecaster
A weekly forecast you can explain: pulls the open pipeline and your own historical stage win rates, computes the forecast in code with a discount for stale deals, gives commit and best-case ranges…
Runs on Salesforce
Smart Site Planning
Plans a greenfield city from its land data and draws the plan on a map: a housing grid on buildable land, water tanks on high ground where pumping is least, the sewage treatment plant on low ground…
Runs on Any platform
Supplier Disruption Impact Analyst
When a supplier or site goes down, walks the supply graph to every part, product and customer order that depends on it, checks stock cover in parallel, quantifies days short and revenue at risk, and…
Runs on NetSuite
Test Automation Generator
Writes the tests a pull request is missing: reads the change and your testing conventions, plans the tests, writes them on a branch, runs them in CI, and comments with the coverage gain - failing…
Runs on Any platform
Test Results Reader
Handles every new lab result: flags values against reference and critical ranges in code, pages the clinician immediately for a critical value, and drafts a plain-language explanation for the patient…
Runs on FHIR EHR
Nothing matches yet. Tell us the process and platform, and we will build it with you: talk to us.
The model
One engine. Five dimensions of knowledge on top.
The engine is the same for every customer. What makes an offer useful is knowledge: of an industry, a process, the people who run it and the system it runs on, delivered at the depth you need.
Which industry
Industry
The entities, terms and rules of one industry.
- Financial services and insurance
- Healthcare and life sciences
- Energy and utilities
- and three more
Which process
Process
How an end-to-end value stream runs, across industries. Based on APQC's process framework.
- Order to cash
- Issue to resolution
- Record to report
- and five more
Who uses it
Team
A crew of agents for one group of people. One process can need several.
- Operations
- Engineering
- Customer support
- Ideation and design
Where it runs
Platform
A native fit to the system of record you already own. Swap it, and the knowledge on top stays the same.
- SAP, NetSuite, Salesforce
- ServiceNow, Jira, Zendesk
- Snowflake, IBM i, FHIR
- and more
How deep
Edition
The same core, used as it is, configured to your rules, or tailored to how you work.
- Ready
- Configured
- Tailored
An offer is one intersection: Order to Cash for Operations on NetSuite, at the Ready edition. We build each intersection once, and offer it on every platform we fit.
What you install
An offer is a working set, not a prompt.
Installing an offer adds its agents, workflows, report and app to your organisation, wired to each other, and asks for a connection to each system it uses: one you have, or a new one.
- Agents with their tools, knowledge and rules
- A workflow that runs the process step by step, in code where the numbers matter
- A report you can run again on next month's data
- An app, pointed at the agents and data that came with it
- Settings asked at install, never guessed: thresholds, codes, approvers
Same offer, every platform
Native on each system. Tested against the same cases.
The Order-to-Cash Revenue Leakage Monitor matches every delivery to its order and invoice, flags what was delivered but not billed, and raises recovery once finance agrees. It runs on NetSuite today. The SAP S/4HANA build reads SAP's own sales, delivery and billing services and raises debit memo requests on billing block.
Each platform is its own build, not a translation layer. Both share one evaluation suite, case for case, and a platform build is listed only once that suite passes against a live system. The SAP build is in that validation now.
Editions
How deep we go is your choice.
Each edition adds a layer around the same core, so you can start small and go deeper without starting again.
Ready
The standard offer, installed as it is, with its settings asked at install. Often the first step.
Configured
Your settings, rules, approvers and data mapping, set at install or by our engineers with you.
Tailored
Logic built only for you, by Relevance Lab's delivery team, on the same core. Often a large enterprise.
Why packaged offers
Buyers pick what they can recognise, price and prove.
Enterprise AI is moving from pilots to programmes. Most pilots never show a return, and most custom agent projects don't survive. Named offers with proof do.
95%
of generative AI pilots showed no measurable profit-and-loss impact.
Packaged offers with proof, not open-ended pilots.MIT NANDA, The GenAI Divide, Aug 2025 · source40%+
of agentic AI projects will be cancelled by the end of 2027, for cost, unclear value or weak risk controls.
Rules per action and a signed record, from day one.Gartner, June 2025 · source70%
of enterprises will abandon agentic AI built for them by vendors' forward-deployed engineers by 2028. Under 20% of those engagements turn recurring needs into product.
Offers built once, and a layer you own.Gartner, September 2026 · source$37B
spent on generative AI by enterprises in 2025, about three times 2024.
Buyers want offers they can buy, not projects.Menlo Ventures, Dec 2025 · sourceAlongside the platforms you own
Hand off, don't replace.
Camaan works across your systems rather than instead of them. An offer fills the gaps in the platform you already run, and agents pass work to and from it, so you keep your investment and your budget stays with the programme it belongs to.
Neutral ground
ServiceNow, SAP or Salesforce is one system among many. Camaan connects it to everything else.
Knowledge, not just hands
Industry and process knowledge that makes agents useful on day one.
A layer you own
Installed offers are copies in your organisation. Change them, extend them, keep them.
Build your own
Partners and in-house teams publish their own offers, described the same way, governed the same way.
Keep ServiceNow for what it does best. Camaan connects it to everything else, and brings your industry's knowledge to it.
Read how we think about packaging agents: One engine, knowledge on top