How agent-ready is PagerDuty?

PagerDuty's agentability audit reveals a mixed readiness for autonomous AI interaction. The homepage and pricing pages scored 45/100, indicating moderate machine-friendly structure, while the documentation scored just 6/100, reflecting significant barriers to programmatic navigation and comprehension. The platform demonstrates strong machine readability (81/100) and control mechanisms (72/100), but falls short on shadow UI avoidance (15/100), defaults (20/100), and transparency (18/100).

The low documentation score stems primarily from poor information chunking and lack of transparency signals that agents rely on to validate and contextualize information. While the underlying content may be technically accurate, its presentation does not facilitate efficient parsing by autonomous systems. The weak transparency and defaults scores indicate that agents struggle to distinguish between authenticated and unauthenticated features, and lack the confidence metadata needed to assess output reliability.

These gaps primarily impact agent-driven workflows such as automated research, integration setup, and troubleshooting. Human users may navigate these issues through contextual understanding, but AI agents require explicit structural cues and metadata that are currently absent or inconsistent across PagerDuty's web properties.

PagerDuty can improve its agentability by implementing the following targeted fixes:

• Add JSON-LD structured data using schema.org vocabularies to help agents identify and extract key entities like products, pricing plans, and API endpoints across all pages.
• Clearly mark which features, documentation sections, and API endpoints are accessible without authentication versus those requiring sign-in, allowing agents to plan workflows appropriately.
• Restructure documentation and FAQ headings as direct questions (e.g., "How do I cancel my subscription?" instead of "Cancellation") to improve agent navigation and intent matching.
• Provide confidence indicators on dynamically generated or aggregated content, such as "verified" badges or certainty scores in API responses, enabling agents to assess reliability.
• Implement a summary-first content structure where each documentation section and important result begins with a 1-2 sentence core answer before expanding into details.
• Pre-populate form inputs with sensible defaults wherever applicable, reducing the decision burden on agents and accelerating automated interactions.