Portal26
AI governance and security platform (AI TRiSM) that discovers AI tools and agents, records agent prompts, responses and tool calls, throttles or stops runaway agents with Agentic Token Control, and answers questions over AI activity with its Recon assistant.
Portal26, formerly Titaniam, sells an AI trust, risk and security management platform as SaaS. It keeps a real time catalog of the public, private and licensed AI tools in use across a company, runs detectors for compliance violations, data exposure and security threats, protects prompts, manages AI policy, and supports audits and forensics of AI use, with modules for license intelligence and AI value measurement.
For AI agents, Portal26 discovers every agent across laptops, hyperscaler environments and SaaS platforms and records each agent's model, users, volume of MCP tool calls and agent to agent calls, conversations and the exact prompts and responses between the agent and its model. Teams can block or remove high risk agents through their existing device management, set policy based limits at the agent, workflow or organization level, and let Agentic Token Control throttle, pause or terminate runaway agents in real time while tracking token spend by agent.
Recon, Portal26's own assistant, answers plain language questions over all of this activity and builds and monitors detection rules. The homepage describes a NIST and SOC 2 certified platform with FIPS 140-2 encryption. It fits security and AI governance teams that need to see and control employee and agent use of AI across the enterprise; its pages do not document an approval step for agent actions, a hosting region, or an API.
Vendor details
Canonical URL
https://portal26.ai
Category
Security / SOC agent
Funding status
Series A. Formerly Titaniam; founded 2019 in San Jose, California, by CEO Arti Raman. The company pivoted from data security and encryption to GenAI trust, risk, and security management.
Company status
independent
Use cases & customers
Primary use cases
Target customers
Integrations
Discovers AI tools and agents across laptops, hyperscaler environments and SaaS platforms, covering assistants such as Claude, ChatGPT, Copilot and Gemini; blocks or removes high risk agents through the customer's existing device management; partners include AWS, Capgemini, HPE, Slalom and Carahsoft. No API or SDK is documented.
Sources & related URLs
Related / legacy domains
Research sources
Agentic Index coverage score
5.5 / 14 capabilities · 39%
| Integrations & Tool Calling | Full |
|---|---|
|
Agents are discovered across laptops, hyperscaler environments and SaaS. Teams can "use their existing device management systems to block or remove high-risk agents," and safeguards "automatically throttle, pause, or terminate runaway agents," so Portal26 acts on the customer's agents through connected systems. The connectors themselves are not listed. Sourceportal26.ai/agentic-ai-security-and-value-realizationread 2026-09-28 |
|
| Workflow Orchestration | Partial |
|
Recon, Portal26's own assistant, answers questions about AI activity and "automatically builds detection rules, monitors their performance," which is a fixed product flow. No flow of several steps that the buyer configures, and no coordination between agents, is documented. Sourceportal26.ai/reconread 2026-09-28 |
|
| Knowledge Grounding & RAG | Full |
|
Recon "turns every AI interaction across your organization into an answer," querying the prompts, usage and interactions Portal26 records across Claude, ChatGPT, Copilot and Gemini, shadow AI included. It is a maintained record of the customer's own AI activity that the assistant retrieves from. Sourceportal26.ai/reconread 2026-09-28 |
|
| Human Oversight & Guardrails | Partial |
|
"Policy-based limits at the agent, workflow, or organisational level" and safeguards that throttle, pause or terminate runaway agents are constraints the customer sets on its own agents. No step that holds an agent action for a person to approve is documented. Sourceportal26.ai/agentic-ai-security-and-value-realizationread 2026-09-28 |
|
| Security, Identity & Governance | Partial |
|
The homepage calls it a "NIST, SOC2-Certified" platform with FIPS 140-2. No SSO, role model or console audit log is documented. Sourceportal26.airead 2026-09-28 |
|
| Observability & Auditability | Full |
|
For each discovered agent, the record holds its model and users, the "volume of tool calls (MCP) and other A2A calls and conversations" and "the exact prompts and responses passing between agent and AI model." The buyer can inspect every exchange of its agents, with AI audit and forensics alongside. Sourceportal26.ai/agentic-ai-security-and-value-realizationread 2026-09-28 |
|
| Memory & State Persistence | Not documented |
|
No memory for Recon with a stated scope and lifetime is documented; the AI activity record is what Recon searches, not memory it keeps. Sourceportal26.airead 2026-09-28 |
|
| Deployment & Data Residency | Not documented |
|
Portal26 is SaaS with a 30 minute activation; no hosting region, region choice or customer environment option is documented. Sourceportal26.airead 2026-09-28 |
|
| Prebuilt Agents / Templates / Packs | Not documented |
|
Portal26 sells governance modules and one assistant (Recon); no prebuilt agents, templates or named policy packs are documented. Sourceportal26.airead 2026-09-28 |
|
| Triggers & Channel Coverage | Full |
|
Detectors watch AI activity in real time, and "adaptive safeguards... automatically throttle, pause, or terminate runaway agents" when limits are crossed, so events start the work without anyone launching it. Sourceportal26.ai/agentic-ai-security-and-value-realizationread 2026-09-28 |
|
| Model Flexibility & Routing | Not documented |
|
Portal26 governs the models its customers use; no model behind Recon is named and no customer choice is documented. Sourceportal26.airead 2026-09-28 |
|
| APIs / SDKs / MCP Extensibility | Not documented |
|
Building on Portal26 is not documented: there is no API, SDK or MCP server for it. It observes the MCP traffic of the customer's agents rather than exposing its own. Sourceportal26.airead 2026-09-28 |
|
| Testing, Debugging & Optimization | Not documented |
|
Audits, forensics and usage benchmarking report on how AI is used. No red teaming or scored testing of the customer's agents or of Recon is documented. Sourceportal26.airead 2026-09-28 |
|
| Browser / Computer-use | Not documented |
|
No agent operating a browser or desktop is documented. Sourceportal26.airead 2026-09-28 |
|
The Agentic Index coverage score grades every vendor Full, Partial or Not documented against the same 14 buyer facing capabilities, from public evidence only. Each capability links to how all vendors in the index score on it. How this evidence is graded
Pricing
Not public. Portal26 is an enterprise AI TRiSM platform sold through a demo and sales motion, with a zero deployment option that turns on through existing security vendors.
not disclosed
Included quota
Not public.
Cost watchouts
Inference, not stated by the vendor: cost may scale with monitored users, AI usage and agent activity; agent token governance may add scope.
Variable cost rationale
Inference, not stated by the vendor: coverage likely scales with monitored users, AI tools and agents; no metering is published.
Overage / add-ons
Not public.
Sales call required
Yes, required for paid access
Free / trial
Schedule a demo.
Lowest paid plan
Not public.
Key ambiguities
No public pricing; enterprise, sales led motion.
Related vendors
- 7AI — Agentic SOC from the Cybereason founders: domain specialized agents…
- Abnormal AI — Behavioral AI email security with three named autonomous agents: a…
- AirMDR — AI-native MDR whose AI analyst investigates every alert and writes…
- Airrived — Agentic OS that unifies SOC, GRC, IAM, vulnerability management, IT,…
- Andesite — Human-AI SOC platform where customers build their own agents and…
- Anvilogic — Agentic security operations platform that works on top of existing…
Alternatives to Portal26
The closest documented capability profiles to Portal26 among security and SOC agents tracked by Agentic Index, ordered by similarity on the same 14 point evidence the rankings use. No vendor pays for placement.
- Linx Security6.5 / 14Adds documented Prebuilt Agents, Templates & Packs
- Clutch Security6.0 / 14Adds documented Deployment & Data Residency and APIs, SDKs & MCP Extensibility
- Pi6.0 / 14Adds documented Memory & State Persistence and Prebuilt Agents, Templates & Packs
- Equixly4.5 / 14Fuller documented coverage on Workflow Orchestration
- Knostic5.5 / 14Adds documented APIs, SDKs & MCP Extensibility and Testing, Debugging & Optimization
- Token Security6.5 / 14Adds documented Prebuilt Agents, Templates & Packs and APIs, SDKs & MCP Extensibility
Similarity is computed from each vendor's Agentic Index coverage score evidence, axis by axis, not from the totals. How this evidence is graded