Glean 拾遗
Week 37 · Thu, Sep 10, 2026

2–3 picks worth reading every day

A bilingual tech zine. Weekly digest on Mondays.

Got something? Submit a link · editor will review.

Today

Today · 2picks

www.kelviq.com · 8 min read

Forgotten Agent Ran 1,555 Sessions a Day and Ate My Claude Quota

The author's Claude Max quota drained in 10 minutes, and he nearly blamed a faulty meter. Digging through Claude Code's local logs, he found a forgotten background agent from a YC batch project spawning 1,555 short-lived sessions a day, with 91% of usage generated by machines rather than by him. The post explains why the limits feel broken: the 5-hour cap is a rolling window that background sessions can already fill, and every fresh session pays a cache-warmup cost roughly 12 times higher than re-reading from cache. The author turned his two-day investigation into tare, an open-source Claude Code skill that deduplicates log entries (naive counts overstate tokens by about 86%) and answers with a cause instead of a raw spreadsheet. It runs entirely locally, makes zero network calls, and can produce a scrubbed HTML usage report that can be shared without leaking prompts or file paths. Useful for anyone whose Claude Code quota keeps disappearing and for teams auditing agent-side spending.

Agent EngineeringAi ToolingClaude CodeCost OptimizationObservability
sreenathmenon.com · 25 min read

WebMCP: Give AI Agents Declared Tools Instead of a DOM to Scrape

WebMCP is an incubation-stage browser standard from the W3C Web Machine Learning Community Group, proposed by Google and Microsoft. Instead of scraping the DOM, a website registers its actions as structured tools with JSON Schema; an AI agent discovers them, calls them with typed arguments, and the page runs real JavaScript inside the user's already-open, authenticated tab. The author validates the pattern with a deployed demo, Career Copilot: it registers 13 WebMCP tools on a single page, and ChatGPT—which now supports WebMCP—used those tools in a real run, aggregating 75 live openings from GitLab, Stripe and Databricks, scoring 24 job descriptions, surfacing skill gaps, shortlisting 4 roles, then stopping for explicit human approval before submitting 3 applications. The post covers the minimal imperative API (one registerTool() call wrapping existing functions), the current Chrome 149 origin trial and flag, the call flow, and the trust model: same-origin restrictions, readOnlyHint / untrustedContentHint annotations, and human-in-the-loop confirmation for consequential actions. It also states limitations honestly: it is early, Chrome-first today, needs site adoption, and the security model is still maturing.

AgentsAI EngineeringMcpWebMCP
This week’s issue

Mondays

#015 · CURRENT

After Code Becomes Cheap, Engineering's True Cost

It has never been cheaper to write code—or easier to overlook what writing code once sheltered us from. This week's selection converges on one uncomfortable conclusion: as AI compresses generation, the real cost of software shifts to everything around it—reading and understanding code, verifying behavior in production, making intent explicit, and deciding who (or what) should act. One strain of articles frames the danger: code factories are producing far more while QA lags, and incident histories show what happens when that gap goes unmanaged. Another strain points to the answer: context engineering, production agent architecture, and delegation skills are replacing line-by-line typing as the core discipline. Beneath it all, tooling for cache hits, forgiving version control, and faster validation exists to keep the human loop tight. This issue is an invitation to rethink where engineering value actually lives once code is no longer scarce.

14 picks 8/31–9/6