Resilinc

Connectivity is described at category level, and the agents act inside the vendor's own portal. No integration catalog or named connector list is published, no custom tool support is described, and the action the agents take, supplier outreach and mitigation workflows, runs through Resilinc's own Supplier Experience portal rather than as authenticated action in the customer's systems. The Azure and Copilot integration is infrastructure rather than an action surface. Ingestion from 150 million-plus public and private sources is the inbound direction.

SourceResilinc, resilinc.ai platform and Supplier Experience pagesread 2026-09-08

Specialized agents autonomously evaluate disruption impact, initiate supplier outreach and identify alternative supply strategies, and the launch releases describe agents orchestrating response workflows spanning supplier outreach, mitigation planning and compliance validation, from detection through to resolution, with mitigation playbooks and one-click WarRooms for supplier engagement.

That is multi-step execution across detection, impact assessment, outreach and mitigation under the vendor's own runtime. The orchestration is described in prose rather than exposed as a configurable workflow surface, and no customer-facing builder is documented.

SourceResilinc, resilinc.ai platform and AI Agents pages, launch releases 2025-05-01 and 2025-07-09read 2026-09-08

Multi-tier mapping maintains a part-to-site map of the customer's supplier network from tier one to tier n, automatically discovering dependencies, critical components and bottlenecks, built on supplier-validated data that suppliers themselves maintain through the Supplier Experience portal, and enriched by the vendor's accumulated disruption intelligence and proprietary R Score.

Agents reason over that maintained structure, which is what lets an event be interpreted against a specific customer's parts and sites rather than reported generically. That makes the map a persistent, maintained and queryable structure over the customer's own network, kept current by the suppliers themselves.

SourceResilinc, resilinc.ai multi-tier mapping, Supplier Experience and platform pagesread 2026-09-08

An approval gate is documented, not merely asserted.

The platform page states that the Ask Resilinc expert delivers proactive alerts and recommended actions and that, with your team's approval, it can initiate mitigation workflows, so the write step is conditioned on a human decision; the governance block names human-in-the-loop as the mechanism ensuring critical judgment sits at the core of operations; and the vendor's own Azure Marketplace blog walks the shape through end to end: agents detect a disruption, map affected parts and identify backup suppliers, after which a supply chain manager reviews the recommendations and launches the contingency plan, always with human oversight.

The May 2025 launch release adds that agents act autonomously only under governed conditions. The autonomy configuration itself, what may run ungated and who sets it, is not documented.

SourceResilinc, resilinc.ai/products/agentic-ai-supply-chain-risk-platform, Azure Marketplace blog 2025-07-09 and launch release 2025-05-01read 2026-09-08

Resilinc holds a strong set of its own attestations, listed on its Data Security page: SOC 2 Type 2, ISO/IEC 27001:2022 with the ISMS scoped to the design, development and operations of its supply chain risk management solutions per its Statement of Applicability, ISO/IEC 27701:2019 for privacy scoped to customer and supplier PII on the Resilinc portal, ISO 9001:2015, Cyber Essentials Plus, and certification under the EU-US, Swiss-US and UK Data Privacy Framework as a processor.

Azure's own security belongs to the host rather than to Resilinc. No SSO, SAML, SCIM, role-based access, permission model, audit-log control or retention setting is documented, and the governance block offers only robust data security, explainable AI and human-in-the-loop. No permission model is published for SC Command or the Supplier Experience portal.

SourceResilinc, resilinc.ai/data-security and /products/agentic-ai-supply-chain-risk-platformread 2026-09-08

The governance block names explainable AI as a platform property, the vendor states that prioritization is not a black box and that agents return personalized summaries with explicit priority rankings so a user can see why an item ranks where it does, and the R Score is published as a proprietary scoring method. That is reasoning shown at the point of decision. No agent run trace, decision log, retention period or export surface is documented anywhere on the site, so there is no record of what an agent did after the fact.

SourceResilinc, resilinc.ai platform page, R Score page and 2026 disruption blog 2026-05-12read 2026-09-08

The Disruption Agent continuously learning from the customer's past actions and supplier responses to give smarter recommendations on each subsequent use is learning absorbed into the model, not a memory layer the customer can reason about. The supplier-validated data and the multi-tier map ground the agents rather than serving as memory. No memory layer is documented, with no scope stated (per user, per supplier relationship, per event), no lifetime or retention given, and nothing describing state an agent carries between runs.

SourceResilinc, resilinc.ai Disruption Agent page and platform pageread 2026-09-08

Resilinc builds the Agentic AI platform on Microsoft Azure and distributes it through the Azure Marketplace, and the ISO 27701 scope refers to the Resilinc portal hosted on the cloud, but no region, region list, customer environment option or selection surface is documented anywhere on the site. The Data Privacy Framework certification and GDPR adherence are legal transfer mechanisms rather than residency.

SourceResilinc, resilinc.ai/data-security and Azure launch release 2025-07-09read 2026-09-08

The vendor publishes a browsable AI Agents section in its primary navigation, with an overview page and a dedicated product page for each agent (the Tariffs Agent, the Forced Labor Compliance Agent and the Disruption Agent), each with its own positioning line, alongside a separately paged capability set (EventWatchAI, Multi-Tier Mapping, RiskShield, Supplier Experience) and three suite tiers.

A buyer can browse the agents and select among them, and each agent addresses a distinct regulatory or disruption domain and is whole without the others. The May 2025 launch release adds mitigation playbooks tailored to disruption types.

SourceResilinc, resilinc.ai AI Agents navigation and product pages, launch release 2025-05-01read 2026-09-08

EventWatchAI tracks global events in real time, and the agents continuously monitor public and private data sources, 150 million-plus by the vendor's own count, for geopolitical, regulatory, cyber and environmental threats, delivering immediate alerts across the supplier network from natural disasters to factory shutdowns and interpreting each event against the customer's own multi-tier map before raising it.

Work fires on continuous external signals rather than user request. Delivery is to the platform, the chat interface and alerting, and no webhook or external channel surface is documented.

SourceResilinc, resilinc.ai EventWatchAI and platform pages, launch release 2025-05-01read 2026-09-08

The vendor's own launch release names the stack plainly: the Agentic AI platform integrates Azure AI Foundry alongside other Azure infrastructure and security services, and the platform is distributed through the Azure Marketplace. No customer or admin model selection, routing control or bring-your-own-model option is documented anywhere on the site.

SourceResilinc, resilinc.ai Azure launch release 2025-07-09 and platform pagesread 2026-09-08

No interface by which an outside caller drives this platform is documented. There is no API reference, SDK, developer portal, MCP server or agent card, no developer or docs subdomain, and the only non-www hosts are the customer portal at portal.resilinc.com behind sign-in, a careers site and a Docebo-hosted academy. No first-party page claims agent-to-agent (A2A) orchestration with other enterprise agents, and no A2A agent card naming an endpoint and auth scheme is published.

SourceResilinc, resilinc.ai estate read, probe recordedread 2026-09-08

No harness, scored test cases, holdout or controlled post-deployment experiment measuring agent decisions is documented, and no configuration or readiness validation gate is described either.

What-if scenario planning tests the customer's supply chain against simulated disruptions using historical and real-time data, revealing vulnerabilities and suggesting alternate strategies, so the artifact under test is the supply chain and the response plan, not the agents; that is the product's own function. The agent improving from past actions and supplier responses is absorbed learning with no documented loop.

SourceResilinc, resilinc.ai/products/agentic-ai-supply-chain-risk-platform and Disruption Agent pageread 2026-09-08

The agents monitor external sources, reason over the multi-tier map and act through the platform's own interface and the Supplier Experience portal; source monitoring at scale is ingestion, not computer use. No browser, desktop or remote control of an interface the vendor does not own is documented.

SourceResilinc, resilinc.ai estate readread 2026-09-08