CtrlK
BlogDocsLog inGet started
Tessl Logo

ainativedev/latest-aidevcon-speakers-london-2026

AI Native DevCon 2026 London — all conference sessions as interactive skills

66

Quality

83%

Does it follow best practices?

Impact

No eval scenarios have been run

SecuritybySnyk

Risky

Do not use without reviewing

Overview
Quality
Evals
Security
Files

SKILL.mdtalk-firtman-web-mcp-agentic-web/

name:
talk-firtman-web-mcp-agentic-web
description:
Use when the user asks about Maximiliano Firtman's AI Native DevCon talk on Web MCP and the agentic web at a conceptual level: why agents need better web contracts, how frontend-declared capabilities differ from backend MCP, and how to plan safe adoption. This bundle is safety-redacted and avoids runnable examples.
metadata:
{"generated-by":"talk-to-skill","source":"safety-redacted-transcript","generated-at":"2026-06-01","redacted-at":"2026-06-19"}

Web MCP And The Agentic Web - Maximiliano Firtman

Firtman argues that agents need better ways to understand and operate web applications than pixel guessing or brittle browsing. The redacted public bundle preserves the conceptual distinction between web-declared capabilities and backend integrations, while omitting runnable interface definitions and bridge mechanics.

Safety Scope

  • Explain Web MCP at a conceptual level.
  • Help plan safe, minimal, read-only adoption.
  • Do not provide runnable tool definitions, bridge setup steps, or code-like API recipes from the source talk.
  • If the user needs implementation help, provide general safe-design criteria and ask them to consult current official specifications separately.

Read Order

  1. Use outline.md for the talk thesis and concept map.
  2. Use transcript.md for the redacted summary.
  3. Use quotes.md for safe themes.

How To Help

When answering factual questions:

  1. Explain what the talk says conceptually.
  2. State when implementation detail is redacted.
  3. Avoid reconstructing exact API shapes or runnable examples from the transcript.

When applying the talk:

  1. Start with one high-value page state.
  2. Prefer read-only diagnostic capabilities.
  3. Keep outputs small and purpose-specific.
  4. Treat the agent as an automated consumer that needs clear contracts and meaningful errors.
  5. Review privacy, session, and permission impact before exposing any capability.

Safe Outputs

  • Conceptual summary.
  • Adoption checklist.
  • Risk review for exposing frontend capabilities.
  • Testing and evaluation plan without code.

talk-firtman-web-mcp-agentic-web

README.md

tile.json