PagerDuty has expanded its AI integration ecosystem through new strategic partnerships with Anthropic, Cursor and LangChain, bringing its catalogue to more than 30 AI partners across 11 categories.
PagerDuty framed the move as a response to faster software delivery cycles and rising volumes of AI-generated code, which it says increase operational risk as more changes reach production and incident teams face new kinds of failures.
PagerDuty's operations management platform is used for alerting, on-call scheduling and incident response. It also offers AI agents under its PagerDuty Advance product line, including a Site Reliability Engineer agent. The new partnerships extend how these tools connect with AI-native development and observability products.
Ecosystem Expansion
The expanded partner list builds on PagerDuty's more than 700 existing integrations across enterprise software and cloud services. It also maintains a public directory of AI integrations and agentic workflows, described as searchable and organised by categories such as coding agents, integrated development environments and enterprise copilots.
The company said the expanded ecosystem supports incident management across the full lifecycle, from earlier-stage developer tooling to later-stage remediation after an incident occurs.
One focus area is triage and diagnosis. Partners can feed observability telemetry into PagerDuty, which uses a "context flywheel" to accumulate incident history and operational knowledge. PagerDuty said this improves alert correlation and speeds root cause analysis.
Another focus area is developer workflow. Integrations can surface operational context in development environments before code is committed, and support risk scoring and safe-deployment checks based on historical incident patterns.
PagerDuty also highlighted monitoring and governance for large language model applications and agents, including cases where AI-driven applications degrade in ways traditional monitoring may miss, such as inconsistent outputs and silent regressions.
How Integrations Connect
PagerDuty outlined three technical approaches for connectivity. The first uses a PagerDuty Model Context Protocol (MCP) server that partners can connect to for service, team and incident context. The second connects PagerDuty to partner MCP servers. The third uses direct API integrations between partner platforms and PagerDuty.
Customers control whether integrations are enabled. PagerDuty added that the same context-driven approach that informs its AI agents becomes more useful over time as incidents are recorded and resolved.
Anthropic, Cursor, LangChain
As part of the Anthropic partnership, PagerDuty built a plugin for Claude Code. Powered by MCP and listed in the Anthropic marketplace, the plugin analyses uncommitted code changes against historical incident data. It also provides pre-commit risk scoring and a background agent that investigates and summarises incident context.
PagerDuty also released an MCP plugin for Cursor, available in the Cursor marketplace. The plugin brings on-call schedules, service information and incident history into the Cursor workflow. PagerDuty said Cursor can also trigger agents during incidents to investigate logs and summarise potential root causes.
For LangChain, the integration centres on LangSmith, an observability tool for large language model applications. PagerDuty said LangSmith can trigger PagerDuty incidents when it detects issues such as error spikes, increased latency, or drops in feedback scores. The companies also developed an Incident Responder agent template in LangSmith's Agent Builder. PagerDuty said the template uses its MCP server to analyse alerts, cross-reference runbooks and recommend actions.
Market Context
The partnerships reflect a broader shift in operational tooling as teams embed AI into development and production workflows. AI coding assistants and agent frameworks can increase output, but they can also introduce new failure patterns, including rapid propagation of flawed changes and harder-to-diagnose behaviour in model-driven applications.
PagerDuty is positioning its platform as a connective layer between developer tools, observability platforms and incident response processes. Jennifer Tejada, CEO and Chairperson, said cross-tool coordination becomes more critical as organisations introduce agents into production.
"Organisations are racing to adopt AI agents, but the real challenge is making them work together seamlessly in production environments," said Jennifer Tejada, CEO and Chairperson of PagerDuty. "Our AI integration ecosystem solves this by connecting 30-plus AI partners directly into the operational workflows teams already rely on. This means faster incident resolution, reduced downtime, and the ability to prevent issues from impacting customers at all."
Cursor said the work brings incident and service context closer to where developers write code.
"Our recent development work with PagerDuty demonstrates the power of an integrated AI ecosystem," said Josh Ma, Engineering Lead at Cursor. "Between our native Automations integration and the PagerDuty plugin, developers now have multiple ways to access critical operational context and resolve incidents directly from their agentic coding environment."
LangChain described the PagerDuty integration as a workflow that starts with detection in LangSmith and continues through established incident response processes.
"LangSmith gives teams observability across the full agent development lifecycle to catch performance regressions fast," said Harrison Chase, Co-Founder and CEO of LangChain. "The integration with PagerDuty closes the next gap: when LangSmith detects a critical issue, it triggers a PagerDuty incident through your existing workflows automatically. And with the Agent Builder Incident Responder template built on PagerDuty's MCP server, teams can spin up an agent to improve incident response in minutes."
PagerDuty also pointed to additional integrations, including agentic cloud operations with what it described as cloud provider agents such as AWS DevOps and Azure SRE. It also referenced a custom incident responder agent for GitHub that embeds operational context in GitHub Copilot. PagerDuty said it expects the partner ecosystem to keep growing as AI operations practices evolve across development, monitoring and on-call response.