Aisera

Breadth plus authenticated action plus runtime tool discovery. Aisera connects to more than a hundred enterprise applications with native one-click integrations to systems including ServiceNow, Salesforce and Jira, and ships a library of over a thousand prebuilt actions covering create, read, update and delete operations, so agents write to systems of record rather than only reading them.

The documentation adds that the platform automatically discovers and configures available tools, and that agents query multiple platforms for data validation within a single diagnostic flow. Ingestion APIs and webhooks are published for teams that would rather not use the admin interface.

Sourcedocs.aisera.com/aisera-platform/llm-operations/understanding-llm-capabilities/aiseras-agentic-ai-for-itsmread 2026-09-05

Multi-step execution with documented decision machinery rather than a single reasoning pass.

Aisera Unify orchestrates Universal, Domain Specific and Task agents that plan, reason and act across entire processes, and the documentation names the mechanisms: Multi-Criteria Decision Analysis evaluating paths on success probability, execution time and resource requirements; Ensemble Decision Methods combining reasoning approaches for consensus; Confidence Scoring with Decision Trees using confidence intervals and branching logic; and Historical Pattern Matching over past interaction data. Hyperflows carry packaged multi-step automations, and the platform automatically discovers and configures available tools at runtime.

Sourcedocs.aisera.com/aisera-platform/llm-operations/understanding-llm-capabilities/aiseras-agentic-ai-for-itsmread 2026-09-05

A maintained retrieval structure documented as a pipeline rather than asserted as a feature: data sources are configured and ingested on a schedule or on demand, PDF, HTML, Markdown, PowerPoint, text and Box notes are converted through the Docling document converter with optional OCR and accurate table parsing, smart caching skips documents whose content has not changed, and an Indexer job must complete before AI Learning or Content Generation can serve from the content.

Documents can be retired from the pipeline so they stop serving user requests while remaining in the source, and custom fields are mapped from source schema into the platform. Aisera's knowledge graph and ontologies are named in the security documentation.

Sourcedocs.aisera.com/aisera-platform/adding-data-to-your-tenant/data-ingestionread 2026-09-05

The TRAPS documentation describes guardrails enforced at runtime rather than a governance essay: a Content Moderation Service validating and sanitizing input at the ingress gateway and inside the service, system prompts fixed at build time instead of free-form input, out-of-distribution detection bounding agents to registered topics and domains, and a human curation step where named roles review any data before it enters model training.

Human confirmation loops that halt high-stakes actions and require approval are described on the vendor's blog rather than in the product documentation; the loops are the part that carries Full, and they are documented at that weaker tier. PII anonymization and injection defense are credited on Security and are not counted again here.

Sourcedocs.aisera.com/aisera-platform/tenant-setup/security-and-privacy-trapsread 2026-09-05

Compliance posture and access surface both documented first-party, with a trust center at trust.aisera.com.

Compliance: SOC 2 Type II audited yearly against the AICPA Trust Service Criteria for Security, Availability and Confidentiality with a risk management strategy submitted, ISO/IEC 27001 certification, Cloud Security Alliance STAR Level 1 certification for the SaaS, GDPR controls, CCPA compliance, and HIPAA BAAs signed on request.

Access: the product documentation carries a role-based access control overview, built-in user roles and permissions, platform permission definitions, user account and role management, and an authentication section.

Sourceaisera.com/platform/security-and-complianceread 2026-09-05

Reconstruction rather than reporting, documented as three named surfaces: an Audit Trail whose logs capture all AI agent decisions with timestamps and reasoning, AI Lens providing complete stack trace visibility into agent conversations and decision-making, and AI Workbench analyzing agent performance and unresolved conversations. The platform additionally states full observability through OpenTelemetry integration within the LLM gateway, which counts as customer capability because the customer receives the telemetry directly rather than the vendor merely using a tool.

Sourcedocs.aisera.com/aisera-platform/llm-operations/understanding-llm-capabilities/aiseras-agentic-ai-for-itsmread 2026-09-05

Not established either way.

The ITSM documentation instructs agents to maintain conversation context across multiple diagnostic rounds, which is within-session state; Historical Pattern Matching leverages past interaction data to inform decisions, which is learning absorbed into an optimization model rather than a memory layer; and the Test Suite replays audited historical conversations, which is a stored transcript corpus, and a store is not a memory layer whoever writes it.

None of those establishes a memory a buyer can name, scope or delete, and none establishes its absence either. The Context Management and Conversations tenant settings named in the documentation index, and trust.aisera.com, are where a buyer should look. Aisera ITSM agentic documentation, Test Suite documentation and documentation index 2026-09-05

Not established either way. No region menu, VPC, single-tenant or on-premises option appears on any first-party page reached, and the security and compliance page does not address residency, so no deployment option can be credited and none can be ruled out. The product documentation points to trust.aisera.com for data handling specifics, and that page was not reached; it is where a buyer should look first.

Sourceaisera.comread 2026-09-05

Packaged assets the customer adopts by route rather than assembles from nothing: a library of prebuilt Domain Specific Agents for IT, human resources and customer service, Hyperflows as packaged multi-step automations documented in the product documentation and validated through a Test and Debug feature, and over a thousand prebuilt actions teams can search, clone and modify.

No-code, low-code and pro-code authoring sits alongside for the cases the library does not cover. A browsable catalog surface was not itself reached, and the published bar treats a catalog as evidence rather than as the definition.

Sourcedocs.aisera.com/aisera-platform/llm-operations/understanding-llm-capabilities/aiseras-agentic-ai-for-itsmread 2026-09-05

The agent both receives and initiates. Work arrives across Microsoft Teams, email, web portals and chat, and Aisera publishes proactive triggers in which agents autonomously detect and remediate major incidents before they cause downtime, which is action taken without a user request. The ITSM documentation shows the same shape operationally, with agents escalating automatically when critical thresholds are exceeded on response time or error rate rather than waiting to be asked.

Sourcedocs.aisera.com/aisera-platform/llm-operations/understanding-llm-capabilities/aiseras-agentic-ai-for-itsmread 2026-09-05

The customer chooses, which is the Full line on this axis. Aisera routes each query to the best-fit large language model and supports its own AiseraLLMs, third-party models, or a model the enterprise brings itself, with the routing itself performed inside a named LLM gateway that also carries the OpenTelemetry instrumentation. Both halves the axis distinguishes are therefore present: vendor-side routing across models, and customer-side selection including bring-your-own.

Sourceaisera.com/platformread 2026-09-05

The product documentation names Ingestion APIs and webhooks as the programmatic alternative to configuring data sources through the admin interface, so the platform itself is addressable rather than only reachable through its own console, and the tenant configuration surface exposes a Tools section and a Federation Service. Native Model Context Protocol, agent-to-agent messaging and AGNTCY support are claimed on the platform page; the grade rests on the documented ingestion APIs and webhooks alone.

Sourcedocs.aisera.com/aisera-platform/adding-data-to-your-tenant/data-ingestionread 2026-09-05

A regression suite with a readable comparable result, which is the Full requirement. The documented Test Suite replays entire audited conversations and phrases generated from ingested knowledge base articles rather than single utterances, and compares current bot responses against historical ones in a single click.

The verdict is explicit and comparable: each case returns Matched, meaning the response is consistent with historical data, or Unmatched, meaning it has deviated and a potential regression is flagged, with a Matching Score attached and an Execution Chat Simulation interface for inspecting unmatched cases.

Alongside it the documentation describes a Test and Debug feature validating Hyperflows against simulated scenarios before production deployment, and the platform states it continuously evaluates and optimizes agent performance through the test suite. No published axis requires a named evaluation framework, and Aisera documents one anyway.

Sourcedocs.aisera.com/aisera-platform/ai-workbench-optimize-conversations/test-suiteread 2026-09-05

No browser, desktop session or remote computer control appears in the product documentation. Aisera's agents reach other systems through integrations, ingestion APIs and webhooks, and the documentation describes models deployed in secure environments behind access controls rather than any agent driving an interface. That is a positive statement about the mechanism, not a silence.

Sourcedocs.aisera.com/aisera-platform/tenant-setup/security-and-privacy-trapsread 2026-09-05