Wayfound

A single line of code connects Wayfound to agents built in-house, with external tools, or purchased from third parties, including Intercom Fin, Qualified Piper and Salesforce Agentforce, and Wayfound states it works with all types of AI agents and workflows and with any agent framework or orchestration layer. Its own Model Context Protocol server and a Salesforce Agentforce application are documented.

All named integrations are agents or agent platforms that Wayfound supervises; no connector through which Wayfound acts on a business system such as CRM, warehouse, ticketing or messaging is documented. A broader integration list appears in Wayfound's own guardian-agent roundup, which ranks vendors including itself.

Sourcewayfound.ai home, faq and ai-agent-supervision-platform pages, docs.wayfound.airead 2026-08-31

The vendor documents a Meetings feature allowing supervised agents to share context, collaborate and learn from one another, and timeline views that pinpoint which agent failed at which stage of a multi-agent process. Closed-loop self-improvement generates new directives and knowledge that can be applied automatically through the MCP server.

The vendor explicitly positions the platform as supervising agents rather than building or executing them, and states it works with any agent framework or orchestration layer rather than replacing one. No task planning, decomposition, tool sequencing or end-to-end workflow execution by Wayfound itself is documented.

Sourcewayfound.ai agentic-workflows, guardian-agents and home pages, docs.wayfound.airead 2026-08-31

For supervised agents, Wayfound identifies knowledge gaps and generates new directives and knowledge to improve their compliance and performance, applied back through the MCP server, which produces grounding material for other agents rather than grounding its own execution. Wayfound's own inputs are agent interaction transcripts and the customer's business guidelines, used as the standard for evaluation rather than as a corpus to answer from. No retrieval-augmented generation, document index, ingestion pipeline, corpus or citation-grounded output by Wayfound itself is documented.

Sourcewayfound.ai faq, home and agentic-workflows pages, docs.wayfound.airead 2026-08-31

The MCP server's evaluate_session tool lets a customer's agentic system check a session for compliance and errors mid-workflow, for example before an invoice email is sent, and the supervision platform analyzes every interaction against the customer's guidelines and sends push notifications on rule breaches, drift and data leaks.

Those are an automated evaluation gate the customer wires into its own agent and after-the-fact alerting. The supervision platform page describes no human approval step, reviewer queue or hold on an action pending a person's decision, and Wayfound's claim that approvals run thirty percent faster describes approvers in the customer's own processes.

Sourcewayfound.ai supervision platform page, docs.wayfound.ai MCP server overviewread 2026-09-29

Wayfound states SOC 2 Type II compliance, enterprise-grade encryption, and a guarantee that agent data is never shared or used for model training. No single sign-on, role-based access control, provisioning, retention configuration or audit logging of user actions within the Wayfound platform itself is documented.

Sourcewayfound.ai pricing and faq pages, docs.wayfound.airead 2026-08-31

The vendor documents capturing and analyzing every interaction of every supervised agent, connecting to agents built in-house, with external tools, or purchased from third parties including Intercom Fin, Qualified Piper and Salesforce Agentforce, with a single line of code.

Documented capabilities include live agent behavior scorecards, root cause transparency into why an agent made a given choice, timeline views identifying which agent failed at which stage, end-to-end transcript supervision, and red, amber or green scoring of every interaction against the customer's own business guidelines.

The vendor describes itself as focused on observability, explainability and alerting for enterprise-scale AI, and OpenTelemetry appears among its integrations.

Sourcewayfound.ai ai-agent-supervision-platform, home and faq pages, MCP announcementread 2026-08-31

Historical agent performance data is kept across supervised sessions and underpins drift detection. Through closed-loop self-improvement, evaluation findings generate new directives and knowledge that can be applied automatically through the MCP server so agents improve over time. A Meetings feature lets supervised agents share context and learn from one another. No memory store, retention period, or mechanism by which a directive derived from one session is selected and applied in another is documented.

Sourcewayfound.ai home, agentic-workflows and guardian-agents pages, docs.wayfound.ai mcp-server overviewread 2026-08-31

Cloud delivery is standard, with on-premises deployment available on request for specific use cases. No virtual private cloud option, customer-managed encryption keys, region selection or detail of on-premises requirements, supported infrastructure or existing deployments is published. Pricing is self-serve, starting at $149 a month on the Innovator plan, consistent with cloud delivery as the default.

Sourcewayfound.ai pricing and faq pages, docs.wayfound.airead 2026-08-31

The vendor supervises agents built in-house, with external tools or purchased from third parties, and positions itself as not building or executing agents. Evaluation frameworks and an OKR structure with performance reviews are documented, but the vendor emphasizes that business users write the evaluation criteria themselves in natural language, so the assessment content is customer-authored rather than vendor-supplied. No template library, prebuilt agent set, shareable configuration, starter guideline pack or catalog is published.

Sourcewayfound.ai faq, home and agentic-workflows pages, docs.wayfound.airead 2026-08-31

Every agent interaction is analyzed automatically as supervised agents run, Wayfound says, with push notifications when issues such as rule breaches or data leaks arise, so its supervisor wakes on the arrival of each supervised session with no person initiating the run. An agentic system can also invoke it mid-workflow through the MCP server. Customer-facing channels do not apply to a supervisor. The ingestion mechanism is described rather than documented.

Sourcewayfound.ai supervision platform page, docs.wayfound.ai MCP server overviewread 2026-09-29

On the Enterprise plan, the pricing page offers Bring Your Own Key, with additional sessions at no charge when the customer supplies its own key, which gives the customer control over the model account behind Wayfound's analysis. That analysis runs on OpenAI and Anthropic models under enterprise agreements, but which makers a customer key may target is not stated, and no model picker or routing control is documented. Wayfound's model agnostic claim refers to the agents it supervises, which may be built on any LLM.

Sourcewayfound.ai pricing page, 31 August read of the FAQread 2026-09-29

Wayfound publishes its own Model Context Protocol server with a dedicated documentation section, exposing named tools including evaluate_session for evaluating session data for compliance and errors directly within agentic workflows, retrieval of a complete list of all agents supervised for an organization, and automatic suggestion of improvements based on ongoing session analysis.

An application programming interface and a one-line-of-code integration are documented alongside a Salesforce Agentforce application. The vendor documents using the MCP server from within a development workflow so engineers iterating on an agent have its real-world performance context available.

Sourcedocs.wayfound.ai mcp-server overview, wayfound.ai home, faq and MCP announcementread 2026-08-31

The vendor documents reading and scoring every agent interaction against business guidelines using foundation models as judge, with business users writing evaluation criteria in plain natural language, and red, amber or green scoring per interaction.

An OKR framework with performance reviews delivers feedback on satisfaction and guideline adherence, with drift detection and action recommendations from ongoing session analysis. Improvement suggestions generate new directives and knowledge and can be applied automatically through the MCP server, closing the improvement loop. The MCP evaluate_session tool exposes evaluation to external agentic systems. Evaluation addresses production behavior rather than pre-deployment testing.

Sourcewayfound.ai agentic-workflows and guardian-agents pages, docs.wayfound.ai mcp-server overviewread 2026-08-31

The vendor describes Wayfound as an observability, explainability and alerting platform that captures and analyses the interactions of agents built elsewhere, connected through a one-line-of-code integration, an application programming interface, or its Model Context Protocol server.

Outputs are scores, alerts, assessments, performance reviews and improvement directives delivered to a dashboard, to human reviewers, or back to supervised agents through MCP, all programmatic interfaces. No browser control, navigation, form filling, screen interaction or desktop automation by Wayfound is documented. Agents that Wayfound supervises may themselves perform computer use, but Wayfound itself does not.

Sourcewayfound.ai ai-agent-supervision-platform, home and faq pages, docs.wayfound.airead 2026-08-31