Sign InOpen Brain
Weekly · August 10, 2026 · Issue #005

feed7 Weekly #005

Six practical upgrades for stronger agent workflows, sharper evals, safer tools, and better context.

Open MarkdownOpen JSONSubscribe via RSS
Get the weekly seven in your inbox
Source-backed AI Signals, small enough to finish and complete enough to carry into your agent.
GitHubGitHub Repo · Needs Review
addyosmani/agent-skills
This pack turns common engineering practices into portable coding-agent workflows for specs, TDD, review and shipping. Its useful idea is to require evidence at each gate, not merely better prompts.FEED7 SUMMARY
agent#skills
arXivPaper · Needs Review
The Bitter Lesson of Tool Calling
Across BFCL v4, models usually handled tools as typed Python calls at least as well as native JSON, suggesting code-based orchestration is worth testing for capable coding agents.FEED7 SUMMARY
benchmark#tool-use
Aug 6, 2026
CursorEngineering Post · Official Source
How Cursor Router chooses the right model for the task
Cursor Router learns task complexity and model fit from production behavior, showing why agent routing should include correction signals, cache costs, and per-task performance.FEED7 SUMMARY
agent#model-selection
Aug 6, 2026
VercelEngineering Post · Official Source
Pause workflows for approval with Chat SDK
Chat SDK can pause a workflow for a verified human decision and resume after seconds or days, without a custom approvals table, action handler, or polling loop.FEED7 SUMMARY
tools#agent-sdks
Aug 6, 2026
What to Ignore This Week
New ways to learn and teach with ChatGPT Work and Codex. The announcement names education plugins but provides no plugin list, capabilities, pricing, availability, or integration details.
Open BrainGive It to Your Agent
THIS WEEK’S SAMPLE CONTEXT
# feed7 Weekly #005 — Context

Use this for: preparing an agent-workflow planning session

- addyosmani/agent-skills (GitHub, needs review)
- Benchmarking Coding Agents on New vs Legacy Codebases — Denys Linkov, Wisedocs (AI Engineer, source linked)
- Learning When to Trust via Selective Context Preference Optimization (arXiv, needs review)