BETA · GAMMA-Ray is published for review. The rules, the mapping and every rating here are experimental and will change. How a rating is made · what exists · send a correction
GAMMA-Rayv0.0.155 · beta

Noxus

GAMMA-Ray report · Spectrum 0.2 beta · Public tier · read 2026-11-10 · not reviewed by the company yet

Noxus sells a platform on which companies build AI agents, which it calls AI Workers or Co-workers, and automated flows that carry out back-office processes in the companies' own systems.

1solution carded
of 6 found
22sources captured
and hashed
70provisions lit
in 8 of 8 instruments
4examples of the agent
at work, with sources

Built only from Noxus's own public pages and, for company basics only, the news sources named beside each fact. Noxus has not seen this report. Nothing was tested and no credentials were used. If you are Noxus and something here is wrong or missing, the correction will be made and dated, and you can claim the card to add evidence.

The agentic solutions

One card per solution. Each level is computed from the evidence by a published rule; a dimension with nothing public reads "not yet evidenced", which is not a no. A card rates a product as documented, not anyone's deployment of it.

Noxus AI Workers (AI Co-workers; 'Agents' in the documentation)

Configurable agents that use flows, integration actions, knowledge bases, tables, MCP servers, code execution and a sandbox to carry out tasks, deployed through the web app, API, SDK, chat widget, Slack, Teams, Google Chat, WhatsApp or Telegram.

Noxus AI Workersa product, as documented
SPECTRUM 0.2 BETA · Public tier · 2026-11-10
A4AutonomyOrchestrateswith options on · as shipped A3
π Moderate · S1·R5·A1 ×3
R3ReachOpen protocol
π Moderate · S1·R5·A1 ×3
C2ControlConfigurablewith options on · as shipped C1
π Moderate · S1·R5·A1 ×3
O2OpennessDocumented
π Moderate · S1·R5·A1 ×3
DraftStage: Scaling (11 of 16)In productionOpen protocol

Standout. Autonomy 4 (Orchestrates): "Create Google Drive Sharing Link or Delete OneDrive File"

Next unlock. Control 3 (Observable): public evidence that it is monitored in operation would get there.

Trust. FACT-Ray: 25 facts: 25 π Moderate · found by an agent, captured and hashed, checked by 2 other agents (same model family, shared repository); no human has signed it.

Standards constellation. 70 provisions lit across 8 of 8 instruments.

Evidence: 15 sources, captured and hashed · the card's page has the vault's read key · Not reviewed by its vendor yet · correct or claim this card

The full card, every answer and its evidence · the governance layer

Flows

Visual node-based workflows with triggers, AI nodes and integration nodes. No card yet (generally available): this solution has not been assessed.

Knowledge Bases and Tables

Document stores with semantic search, and structured tables, that agents and flows query. No card yet (generally available): this solution has not been assessed.

Case management and processes

Processes whose cases are worked by flows and by people through tasks. No card yet (turned on per organization): this solution has not been assessed.

Genie

An assistant inside the platform that builds and changes flows, agents and tables, asking for approval before writes. No card yet (beta, off until an administrator turns it on): this solution has not been assessed.

Noxus MCP server, Python SDK and REST API

A hosted MCP server, SDK and REST API that let external clients and agents drive workflows, agents and conversations. No card yet (generally available): this solution has not been assessed.

Our reading

The shape. Noxus AI Workers read A4 · R3 · C2 · O2 (Scaling). A builder configures agents, which the documentation also calls co-workers, that use flows, integration actions, knowledge bases, tables and MCP servers to take actions themselves, and deploys them through the web app, the API, a chat widget or channels such as Slack, Teams and WhatsApp. Each agent reaches only the tools selected for it, an enumerated list.

Where the evidence is strongest. Reach 3 holds as shipped: a hosted MCP server lets external clients and agents call Noxus agents, conversations and workflows, and an OpenAPI 3.1.0 description resolves. Logs record tool calls and reasoning steps, an agent's deployment on a channel can be deactivated through the API, and a public status page posts incidents. Customer stories on Noxus's own pages give the company's figures: at CUF an AI Worker processes over 3,000 patient communications a month and routes complex cases to staff, and LSIG reports 15,000 products a day classified across three retail brands.

Two levels hold with options on. Autonomy 4 and Control 2 rest on the Human in the Loop tool or Require Approval, which are off until turned on; with them, the agent acts on its own and asks a person before critical actions. As shipped, actions run without asking, and the split between routine work and exceptions is shown at CUF, a deployment.

What would move it. Autonomy is at the top of its ladder. Reach 4: agents working across organisations or stacks, with third parties publishing integrations or agents for Noxus. Control 3: a public record of agents' conduct being monitored while they run, such as an alert or a dated review. Openness 3: published evaluation results with a method; Genie evaluations are announced, and an OpenAPI description and a status page are in place.

Seen, and not counted. The challenger lowered monitoring: Agent Insights analyses what users ask and how they feel, not what the agent does, and the Monitoring page covers infrastructure. A sandbox outside the agent that blocks internal network ranges is documented as an option; it cannot lift Control while level 3 is not yet evidenced. Certification claims differ between Noxus pages, and no certificate was read.

What would move it

The one piece of public evidence that would lift each card next:

The agent at work

Examples from public pages that name a customer or give figures. Each source was captured and hashed on the date shown.

CUF (José de Mello Saúde)

Patient messages categorised, routed and processed by an AI Worker; the page gives 3,000+ communications a month and complex cases routed to staff. "Complex cases are flagged and routed to human staff instantly"

Source (captured 2026-11-10T15:47Z, sha256 278999c51d7a72d7)

LSIG (360hyper, KuantoKusta, Saibarato)

Supplier product data classified and enriched by one AI Worker across three retail brands; the page gives 15,000+ products a day. "LSIG automated 15,000 daily SKUs across three retail brands"

Source (captured 2026-11-10T10:39Z, sha256 e013db9a58df6cfd)

Sandock Austral Shipyards

Tender documents assembled, stamped, signed and validated by a Noxus workflow. "cutting preparation time by 75%"

Source (captured 2026-11-10T10:39Z, sha256 438d25297dfa9617)

Kirchhoff

Supplier documents processed into SAP after a Noxus workshop. "They were live in thirty days without touching their SAP core"

Source (captured 2026-11-10T10:39Z, sha256 99b43a7e9770915f)

The company

The people

Public and professional only, from the company's own pages or a public professional source.

The graph

The company, its people and the stories on the left; each solution, the captured sources behind its card, the indicators those sources evidence and the provisions of the instruments those indicators touch, to the right. "Touched" never means "complies".

people and storiescompanysolutions15 sources25 indicators70 provisions, by instrumentI03: Agents and flows are configured in a visual builder without code, and agents search knowledge bases built from the organisation's documents and data.I04: With the Code Execution tool turned on, the agent writes and runs code.I08: Agents use tools and run flows to take actions themselves.I10: The tools page states one agent can be given integration actions as tools, such as Send Email, Create Google Drive Sharing Link or Delete OneDrive File, alongside flows, knowledge bases and database connections.I11: An agent can be bound to external MCP servers and call their tools.I17: At CUF the AI Worker processes patient messages and routes complex cases to staff.I18: The tools page documents a Human in the Loop tool an agent owner may enable, under which the agent escalates to a person to get approval for critical actions; it does not say how any given agent is shipped.I20: The company states the agents' purpose: handling complaints, documents and tickets.I21: The builder selects which actions and flows each agent may call.I22: The tools page documents limits, such as web research being less reliable than knowledge bases.I23: Each agent's tools are an enumerated list chosen by the builder: the flows, actions, knowledge bases, database connections and MCP servers it may use.I25: Each agent reaches only the flows, actions, knowledge bases and database connections selected for it; its SQL is generated read-only and runs with that connection's privileges, and tables are read-only unless edits are allowed.I26: A Google Chat deployment can run under a service account bot identity separate from any person.I38: The Human in the Loop tool and Require Approval make the agent ask a person before actions run.I39: An agent's deployment on a channel can be deactivated through the API.I41: Logs record agent interactions including tool calls and reasoning steps, with run-level logs per run.I45: When the sandbox is on, a Syd sandbox outside the agent blocks internal network ranges and requires the sandboxes:run permission.I47: Instructions written in prose set allowed behaviour and restrict access.I54: A public status page posts incidents and offers incident update subscriptions.I57: With a Phoenix or Langfuse provider connected, users can rate agent responses up or down.I58: The tools page describes the data an agent can use: knowledge bases, connected databases queried read-only, and workspace tables.I61: The model providers and model families an agent can use are listed.S1: A named customer, CUF, reports patient communications processed by the AI Worker each month.S3: An OpenAPI 3.1.0 description of the backend API resolves; the docs llms.txt also resolves.S4: A hosted MCP server lets external clients and agents call Noxus agents, conversations and workflows as tools, authenticated with a workspace API key whose permissions gate the tools.NoxusJoão Pedro AlmeidaJorge PessoaGonçalo FerreiraMiguel RibeiroCUF (José de Mello Saúde)LSIG (360hyper, KuantoKusta, Saibarato)Sandock Austral ShipyardsKirchhoffNoxus AI Workers (AI Co-workers; 'Agents' in the documentation)Noxus AI WorkersA4 · R3 · C2 · O2FlowsFlowsKnowledge Bases and TablesKnowledge Bases and TablesCase management and processesCase management and processesGenieGenieNoxus MCP server, Python SDK and REST APIMCP server, SDK and APIdocs.noxus.ai/core/what-is-noxusdocs.noxus.ai/platform/agents/too…docs.noxus.ai/platform/agents/int…noxus.ai/case-studies/cuf-healthc…noxus.ai/aboutusdocs.noxus.ai/platform/agents/dep…docs.noxus.ai/api-reference/v1--d…docs.noxus.ai/deployment/operatio…docs.noxus.ai/platform/agents/san…docs.noxus.ai/core/concepts/agentsstatus.noxus.aidocs.noxus.ai/core/infrastructure…docs.noxus.ai/core/infrastructure…backend.noxus.ai/openapi.jsondocs.noxus.ai/sdk/mcp/overviewIt is configured on a low-code or no-code builder and acts on real organisational dataI03It generates and executes code with local or cloud privilegesI04It executes actions itselfI08It works across several domains or systemsI10It connects to external tools or services across a trust boundaryI11The agent handles routine tasks; the person is consulted on exceptionsI17The agent works independently; the person approves critical decisions onlyI18Its purpose and permitted tasks are stated in writingI20Its autonomy limits and the tools it may use are definedI21Its limits and known failure cases are documented for those who use itI22What it can reach is enumerated: scopes, permissions, toolsI23It asks for narrower access than the whole accountI25It has its own identity, separate from any person'sI26High-impact actions need a person's approvalI38There is a way to pause or stop itI39Its actions are loggedI41A limit on it is enforced by something outside its own reachI45A limit on it is a rule in proseI47Incidents and errors are communicated to those affectedI54There is a channel to report problems with it or contest its outputI57The data it uses is describedI58The models and third-party components it uses are listedI61It is in production: a public source names a customer or gives usage figures for what it does on its ownS1It is machine-readable for agents: it publishes an llms.txt, an AGENTS.md, an API description or an agent cardS3It exposes an agent interface: a published MCP server, an agent card, or an API documented for agents to callS4Salesforce 14EAAMM 3OWASP 13MCP authorization 6AAGMM 6Microsoft 8NIST AI RMF 15EU AI Act 5OWASP, AT3 Citizen Developer AgentOWASP, AT4 Code Executing AgentSalesforce, Level 2: Simple Orchestration, Single DomainSalesforce, Level 1, how to advance from Level 1 to Level 2Microsoft, 200: Repeatable, Business strategySalesforce, Level 3: Complex Orchestration, Multiple DomainSalesforce, Level 2, how to advance from Level 2 to Level 3Salesforce, Level 3, how to advance from Level 3 to Level 4Microsoft, 400: Capable, Business strategyMicrosoft, 200: Repeatable, Technology and dataOWASP, AT6 Externally Extended AgentNIST AI RMF, GOVERN 6.1NIST AI RMF, MAP 4.1EAAMM, L3 Consultative (survey item Q9)EAAMM, L4 Supervisory (survey item Q9)NIST AI RMF, MAP 1.1NIST AI RMF, MAP 2.1NIST AI RMF, MAP 3.3NIST AI RMF, MAP 1.3EU AI Act, Article 13OWASP, Maturity Level 1 Experimentation without Guardrails, key enterprise actionNIST AI RMF, MAP 2.2Microsoft, 300: Defined, Business strategyNIST AI RMF, MEASURE 2.5MCP authorization, Authorization (draft)MCP authorization, Authorization (draft)AAGMM, GD-02MCP authorization, Authorization (draft)Salesforce, Level 2, how to advance from Level 2 to Level 3Salesforce, Level 3, how to advance from Level 3 to Level 4MCP authorization, Authorization (draft)OWASP, Maturity Level 4 Adaptive, Self-Regulating Governance, key enterprise actionOWASP, Maturity Level 2 Policy-Defined, Human-in-the-Loop, key enterprise actionEU AI Act, Article 14AAGMM, GD-06NIST AI RMF, MAP 3.5Salesforce, Level 3, how to advance from Level 3 to Level 4OWASP, Maturity Level 3 Integrated, Continuous Oversight, key enterprise actionNIST AI RMF, MANAGE 2.4OWASP, Maturity Level 0 Unaware and Ad Hoc, key enterprise actionOWASP, Maturity Level 1 Experimentation without Guardrails, key enterprise actionEU AI Act, Article 12AAGMM, GD-03AAGMM, GD-08OWASP, Maturity Level 3 Integrated, Continuous Oversight, key enterprise actionAAGMM, GD-04Microsoft, 200: Repeatable, AI governance and securityMicrosoft, 200: Repeatable, AI governance and securityMicrosoft, 500: Efficient, AI governance and securityOWASP, Maturity Level 0 Unaware and Ad Hoc, key enterprise actionNIST AI RMF, MANAGE 4.3EU AI Act, Article 73EU AI Act, Article 26NIST AI RMF, MEASURE 3.3NIST AI RMF, GOVERN 5.1Salesforce, Level 1, how to advance from Level 1 to Level 2Salesforce, Level 4, how to maximize valueNIST AI RMF, GOVERN 5.2AAGMM, GD-05EAAMM, Readiness dimension 2Salesforce, Level 0, how companies can advance from Level 0 to Level 1Salesforce, Level 1, how to advance from Level 1 to Level 2OWASP, Maturity Level 2 Policy-Defined, Human-in-the-Loop, key enterprise actionNIST AI RMF, MAP 1.4Microsoft, 400: Capable, Business strategySalesforce, Level 0, how companies can advance from Level 0 to Level 1MCP authorization, Authorization (draft)MCP authorization, Authorization (draft)OWASP, Maturity Level 1 Experimentation without Guardrails, key enterprise actionSalesforce, Level 3, how to advance from Level 3 to Level 4

The evidence

SourceReadSHA-256 (first 16)
backend.noxus.ai/openapi.json2026-11-10T15:47Zacf57bee22546b3a
docs.noxus.ai/api-reference/v1--deployments/deactivate-deployment2026-11-10T15:47Zed6ad19208950bde
docs.noxus.ai/core/concepts/agents2026-11-10T15:47Z274757353a8ab71b
docs.noxus.ai/core/infrastructure/models-providers2026-11-10T15:47Zd0fe4689f42a1a5b
docs.noxus.ai/core/infrastructure/observability-providers2026-11-10T15:47Z93a416e3707f51cb
docs.noxus.ai/core/what-is-noxus2026-11-10T15:47Zfa50eddb7ba2d529
docs.noxus.ai/deployment/operations/logging2026-11-10T15:47Z6170b7205da6babb
docs.noxus.ai/platform/agents/deployments/google-chat2026-11-10T15:47Z450076f7f0a12e45
docs.noxus.ai/platform/agents/introduction2026-11-10T15:47Za775465cbeab0371
docs.noxus.ai/platform/agents/sandbox2026-11-10T15:47Z6b0d907e6d65015c
docs.noxus.ai/platform/agents/tools2026-11-10T15:47Z107a0dcbda489616
docs.noxus.ai/sdk/mcp/overview2026-11-10T15:47Z554ecd31cdfb2baf
docs.noxus.ai/sdk/mcp/tools2026-11-10T10:39Zbf2ad9c2bb60b44f
observador.pt/2025/02/11/noxus-startup-com-adn-portugues-recebe-15-mil2026-11-10T10:39Zd802e15e98ed9188
status.noxus.ai/2026-11-10T15:47Zf2d62204052945e4
www.noxus.ai/2026-11-10T10:39Zb3d589bfcdbc84aa
www.noxus.ai/aboutus2026-11-10T15:47Z0109bcdc540a1a13
www.noxus.ai/blog/sandock-austral-achieves-75-faster-tender-submission2026-11-10T10:39Z438d25297dfa9617
www.noxus.ai/case-studies/cuf-healthcare2026-11-10T15:47Z278999c51d7a72d7
www.noxus.ai/case-studies/lsig-retail2026-11-10T10:39Ze013db9a58df6cfd
www.noxus.ai/kirchhoff2026-11-10T10:39Z99b43a7e9770915f
www.noxus.ai/pricing2026-11-10T10:39Z716e94a62ee89e6d

How this report was made

A researcher-assessor gathered and captured the sources and mapped the evidence to indicators, without assigning a level. A challenger argued against every level three and four and every doubtful credit. The deterministic scorer computed the levels and the gate checked the rules. A card appears here only after two rounds of review and a sign-off by agents that did not write it (the team's workflow); Noxus AI Workers (AI Co-workers; 'Agents' in the documentation): reviewed in 2 rounds, signed off by gamma.approve@web-summit.dinis on 2026-11-10. A person's sign-off of the series is still to come.

RoleWho
researcher-assessora sub-agent of gamma.studio@web-summit.dinis
challengera sub-agent
scorerbuild/spectrum.mjs
analyst, designer, writergamma.studio@web-summit.dinis
reviewgamma.review@web-summit.dinis, 2 rounds
sign-offgamma.approve@web-summit.dinis, 2026-11-10

The vault

Every capture, the dossier, both assessors' mappings, the challenger's notes and this report's data are in a public sgit vault (6n9a4tow). Read key:

sgit_public_read_eafc52a9cbd090f15069febc32ea08d2630b1aa8410e8b8bc4cc5f4b3f197e98:6n9a4tow

With sgit: sgit clone followed by the key. The key reads; it cannot write.

Corrections

None yet. A correction is made in place, dated here, and carried to the card, the vault and the article.

The article

Noxus AI Workers on the GAMMA-Ray Spectrum: A4 · R3 · C2 · O2, written from this report for PrompTugal.news.