Manny Silva (Skyflow): Self-Healing Docs: How Agents are Redefining the Docs Pipeline

This talk was presented at the AI The Docs 2025 online conference. We are thrilled to share the recording and the summary with you. 

Visit the talk summary page to see all of the presentations from the conference.

presentation

Manny Silva (Head of Docs at Skyflow )

Summary

What is "self-healing documentation," and how does agentic AI enable it? 

In his talk, Manny Silva (Head of Docs at Skyflow) explains how agentic AI can transform documentation from high-maintenance systems into self-healing ecosystems. Traditional docs-as-code pipelines (built on git, IDEs, static site generators, and CI/CD) still struggle with manual content creation, time-consuming testing, cascading maintenance issues, and broken docs that erode developer trust. 

Self-healing documentation tackles these problems by enabling systems to automatically detect errors, propose fixes, create pull requests, and even update themselves. Silva highlighted three tool categories powering this shift: 

• Content generation tools (e.g., Verbiage GPT) that rapidly draft technical content.
• Linters to enforce style and formatting.
• Content testing tools (e.g., Doc Detective) that run docs as tests, validating code, APIs, CLIs, and even screenshots. 

In a self-healing flow, a validation tool flags an error, an agent investigates root causes across docs and product systems, then either drafts a fix for review or escalates product bugs. This can run continuously. 

Manny Silva stressed that the goal is not to replace writers but to empower them. By offloading repetitive maintenance, writers can focus on strategy, UX, and innovation, while gaining peace of mind that their content is accurate, trustworthy, and always up-to-date. 

Main takeaway: Self-healing documentation, powered by agentic AI, shifts the role of technical writers from constant maintenance to higher-value work. Instead of chasing broken links, outdated examples, or stale guides, writers can rely on AI-driven systems to keep docs accurate and up to date, freeing them to focus on strategy, UX, and innovation.