Immersive Fusion announces the release of IAPM 1.12.0, the most significant update to the AI Assistant since its debut in September 2025. This release transforms Tessa from a conversational assistant into a multi-agent system with 6 specialized skills, 46 purpose-built agents, and the ability to analyze screenshots alongside text.

Skills: From One Prompt to Forty-Six Specialists

Previous versions of Tessa operated as a single conversational agent. Powerful, but general-purpose. Version 1.12 introduces the Skills system, an extensible architecture that organizes Tessa's capabilities into specialized domains, each with dedicated agents trained for specific types of work.

The six skills shipping in 1.12.0:

Problem Solving (9 agents). Structured problem exploration with specialists for research, root cause analysis, architecture decisions, quality scoring, validation, synthesis, code review, failure investigation, and progress reporting. Each agent follows established methodologies: the investigator applies 5 Whys and Ishikawa diagrams, the architect produces Nygard-format Architecture Decision Records, the critic scores deliverables across six quality dimensions.

Red Team (11 agents). Full MITRE ATT&CK kill chain coverage for security assessment. Scope-gated: an Engagement Lead agent must establish Rules of Engagement before any offensive agent can operate. Three agents (persistence, exfiltration, social engineering) require explicit RoE authorization. This brings professional penetration testing methodology directly into the observability workflow for teams responsible for both uptime and security posture.

Engineering Team (10 agents). Security-first SDLC from architecture through post-deployment incident response. Includes solution architects with STRIDE/DREAD/PASTA threat modeling, backend and frontend security specialists (OWASP Top 10, ASVS 5.0), infrastructure hardening (CIS Benchmarks, SLSA), DevSecOps pipeline configuration (SAST/DAST), security QA, and incident response runbooks.

NASA Systems Engineering (10 agents). NPR 7123.1D process implementation through specialized agents covering requirements engineering, technical architecture, verification and validation, system integration, risk management (NPR 8000.4C), configuration management, formal review gates (SRR, PDR, CDR, FRR), trade space exploration, quality assurance, and SE status reporting.

Adversary (3 agents). Tessa's self-critique mechanism. Maps criticality levels (C1 advisory through C4 mission-critical) to adversarial strategies, executes those strategies against deliverables, and scores quality using LLM-as-Judge methodology across six dimensions with PASS/REVISE/ESCALATE verdicts. The AI assistant critiques its own output before presenting it.

Prompt Engineering (3 agents). A meta-skill for building better prompts, with interactive 5-element prompt assembly, constraint generation, and a 7-criterion quality rubric.

Built on the Jerry Framework

All skills are ported from the Jerry framework, an open-source multi-agent orchestration system created by Adam Nowak. Jerry defines the skill architectures, agent specifications, routing patterns, and quality scoring methodologies. Immersive Fusion ported Jerry's skill definitions into the IAPM assistant library, preserving the original domain expertise, constraint models, and workflow patterns while adapting them for native execution in IAPM's infrastructure. Every ported skill carries provenance headers citing the Jerry source path, version, and license.

Automatic Routing

Users don't need to know that 46 agents exist. The Skills system includes a two-phase routing engine: keyword matching for common queries (instant routing) and LLM-based intent classification as fallback for ambiguous requests. Ask "why is checkout slow?" and Problem Solving activates. Ask "review this for security vulnerabilities" and the Engineering Team responds. Ask "what's the attack surface?" and the Red Team engages (after scope verification). The routing is transparent. Users just ask questions.

Multimodal Vision

IAPM 1.12 adds multimodal vision support. Users can now paste screenshots directly into the AI Assistant using Ctrl+V and ask Tessa to analyze what she sees. A dashboard screenshot, an error dialog, a log output, a topology view: paste it, ask "what's wrong here?", and Tessa combines visual analysis with her telemetry access to produce a diagnosis.

This eliminates the describe-copy-paste-wait cycle that slows down incident response. Instead of typing a description of what you see on another screen, show it directly.

Try It

IAPM 1.12 is available today for all subscription tiers, including Start (free).

To experience the Skills system and multimodal vision, start free and populate your grid with realistic telemetry using our open-source trace generator:

go install github.com/ImmersiveFusion/if-opentelemetry-tracegen/cmd/tracegen@latest
tracegen -endpoint otlp.iapm.app:443 -headers "api-key=YOUR_KEY" -complexity light

Then ask Tessa anything. She'll route to the right specialist.

Watch it in action: From Error to Full RCA in One Conversation | Introduction to Immersive APM

About Immersive Fusion

Immersive Fusion LLC builds Immersive APM (IAPM), an application performance monitoring platform that renders distributed systems in real-time 3D. Founded on the belief that engineers deserve to see their applications the way those applications actually run, Immersive Fusion combines spatial visualization, AI-guided diagnostics, and full OpenTelemetry support to deliver observability that goes beyond dashboards.

Download: immersivefusion.com/download | Steam | Documentation

Media Contact: Dobri Kowalski, dkowalski@immersivefusion.com

Start Free. Immersive. AI-guided. Full-stack observability. Enter the World of Your Application®.