Start Debugging

Tag: llm

74 posts · Page 3 of 8

2026-09-01 comparison github-copilot ai-agents

Copilot Memory vs Repository Custom Instructions vs AGENTS.md: Which One the Model Actually Reads

Instructions files are deterministic and always sent. AGENTS.md sits at the bottom of the documented repository precedence list. Copilot Memory is a separate store Copilot writes for itself, read by only three surfaces, and it is not in the precedence list at all. Here is the support matrix and the rule for deciding where each convention belongs.

2026-09-01 ai-agents github-copilot llm

VS Code 1.135 Ships /rubber-duck, and It Deliberately Uses a Different Model

The experimental /rubber-duck command in VS Code 1.135 hands the agent's plan, code, and tests to a model from another family for review. GPT-5.4 critiques Claude, and the cross-family choice is the whole point.

2026-08-31 comparison agent-skills mcp

Agent Plugins 1.0 vs Vendor-Specific Plugin Formats: What the Shared Standard Actually Covers

Agent Plugins 1.0.0 standardizes exactly two component types: skills and MCP servers. Slash commands, hooks, subagents, rules, LSP servers, permission config and secrets all stay vendor-specific. Here is the field-by-field matrix across Claude Code, Cursor, Copilot and Codex, and which layout to author in.

2026-08-30 mcp ai-agents llm

Stateful vs Stateless MCP Servers: What Actually Breaks When the Session Goes Away

Protocol revision 2026-07-28 deleted Mcp-Session-Id, the initialize handshake, resources/subscribe, ping, logging/setLevel and SSE resumability. An automated survey of 1000 open source MCP servers found 90% reference the session ID nowhere. Here is what the other 10% has to rewrite, with verified code on @modelcontextprotocol/server 2.0.0.

2026-08-21 cursor ai-agents cloud-agents

How to Cut Cursor Cloud Agent Startup Time With Prebuilt Builds

Cursor shipped Builds for Cloud Agents on August 13, 2026: a bootable filesystem snapshot of an already prepared machine. Cursor reports 3x faster time to first token. The lever you control is the install vs start split in .cursor/environment.json.

2026-08-20 mcp agent-skills microsoft-agent-framework

How to Serve Agent Skills from an MCP Server in .NET with UseMcpSkills

Stop shipping SKILL.md folders inside every agent deployment. Serve them from an MCP server and pull them with UseMcpSkills on Microsoft.Agents.AI.Mcp 1.18. Includes the exact skill://index.json shape, a working C# server, the wire trace, and the SEP-2640 drift that will bite you.

2026-08-19 github-copilot ai-agents llm

How to Set the Reasoning Level for a GitHub Copilot Cloud Agent Per Task

Since August 3, 2026 you can pick a reasoning level next to the model when you start a Copilot cloud agent task. The Agent Tasks REST API has no equivalent parameter, so scripted dispatch cannot set it. Here is where the control exists, which levels each model exposes, what lands on the wire, and how to pin a default instead.

2026-08-17 agent-skills mcp claude-code

How to Package Skills and an MCP Server as One Agent Plugin

Agent Plugins 1.0 says plugin.json at the root and mcp.json beside it. Claude Code 2.1.x wants .claude-plugin/plugin.json and .mcp.json. Here is what each client actually loads, tested against a real plugin, and the dual-layout directory that satisfies both.

2026-08-15 mcp ai-agents llm

Safe File-Write Tools for an Agent: Preview, Confirm, Apply

A write tool that asks before it writes needs three things the MCP SDK does not give you: an approval bound to the exact diff, a precondition on the file it previewed, and an integrity-protected requestState. Verified end to end on @modelcontextprotocol/server 2.0.0 against protocol revision 2026-07-28, including the argument-swap that gets past a naive confirm.

2026-08-14 mcp ai-agents llm

MCP Server Design for a Large Internal API Surface

Mapping 200 internal REST endpoints to 200 MCP tools puts 200 KB of JSON schema in front of every request. The 2026-07-28 spec also made the usual escape hatch illegal: tools/list MUST NOT vary per-connection, so per-session dynamic registration is gone. Here are the three levers that still work, with measured wire sizes and verified code on @modelcontextprotocol/server 2.0.0.