AI & Machine Learning10 min
Salesforce Will Meter AI Agent Calls: What to Do
Salesforce plans to charge Flex Credits for each successful call a registered AI agent makes over MCP or API. Who is affected, when, and what to do now.
Technical Blog
Blog
Expert tutorials & guides for modern developers
685 posts · Page 2 of 58
AI & Machine Learning17 min
MCP Token Audience Check: expectedResource Tested (2026)
I tested the MCP TypeScript SDK's new expectedResource option for token audience validation. It is off by default and Express adapter 2.0.1 ignores it (2026).
Security & Privacy13 min
Claude Code Mod to Block .env Reads: Tested (2026)
Build a Claude Code mod to block .env reads. It denies Read, Edit, Write and Bash calls on secret files, passes 5 tests, and leaks on 5 of 14 shell commands.
AI & Machine Learning18 min
Restate Durable Agents: Duplicate Tool Calls Tested (2026)
I ran nine crash tests on a Restate agent. One repeated a charge, one repeated a paid LLM call, and an idempotency key fixed the charge. TypeScript, 2026.
AI & Machine Learning16 min
AG-UI 1.0 Migration: What Breaks, Tested (2026)
AG-UI 1.0 migration, tested: what a 1.0.1 client does with 0.x streams, the multimodal tool result that crashed a 0.x client, and a protocolVersion fix.
AI & Machine Learning37 min
MCP Events Tutorial: Build a Webhook Server (2026)
MCP Events tutorial: a tested Node.js webhook server with events/subscribe, Standard Webhooks signing and challenge checks. The SDKs I checked lack it.
AI & Machine Learning35 min
Google ADK 2.0 Workflow: Retries and Timeouts Tested (2026)
ADK 2.0 Workflow retries are jittered, unmatched routes raise no error, and timeouts can't stop blocking code. Tested: google-adk 2.10.0, Python 3.10-3.13.
LLM & Integration19 min
Claude Inline Tools Beta: What Actually Ships (2026)
Claude's inline tools beta adds tools mid-conversation without busting the prompt cache. I traced exactly when each SDK shipped it, wire format included.
LLM & Integration17 min
Claude Opus 5.5 tool_choice 400: 3 SDKs Tested (2026)
Claude Opus 5.5 rejects forced tool_choice with a 400. I captured what the AI SDK, LangChain and the raw SDK actually send, and built a preflight check.
AI & Machine Learning13 min
OpenTelemetry gen_ai Attributes: 60 Deprecated in 2026
All 60 gen_ai attributes in OpenTelemetry's core semantic conventions are now deprecated. I audited every release: 8 renamed, 2 removed, 50 simply moved.
Security & Privacy16 min
AI SDK Tool Drift Detection: What It Misses in 2026
I tested AI SDK v7 fingerprintTools and detectToolDrift against a real MCP server. It catches description and schema rug pulls, but not annotation drift.
AI & Machine Learning14 min
MCP Tasks Extension: A Working Python Server (2026)
The ratified MCP Tasks extension needs three new methods. The official Python SDK v2.2.0 still doesn't implement them. Here's a tested server that does.
