VQV Signal

AI AT WORK WATCH TECHNICAL

Real Python thread blames context window for coding agents forgetting rules

According to a post on X by @realpython, Real Python published a thread that blames the context window when a coding agent forgets a rule such as "use uv." The thread lists five failures, each with a suggested fix: forgotten rules, made-up functions, rerunning a failing test, picking the wrong MCP tool, and two rules that disagree. The available excerpt is truncated, so the full fixes are not shown here.

Source: Scobleizer News Pulse (X) · x.com Found via Robert Scoble’s Scobleizer News Pulse · a post on X, not independently verified Published 2026-11-20T23:55:00+00:00 Detected 2026-11-21T05:01:56+00:00
View original source

According to a post on X by @realpython, Real Python published a thread that blames the context window when a coding agent forgets a rule such as "use uv." The thread lists five failures, each with a suggested fix: forgotten rules, made-up functions, rerunning a failing test, picking the wrong MCP tool, and two rules that disagree. The available excerpt is truncated, so the full fixes are not shown here.

AI-assisted summary based on the listed source.

Real Python posted a thread blaming the context window when a coding agent forgets a rule like "use uv". It lists five failures with a fix for each: forgotten rules, made-up functions, rerunning a failing test, picking the wrong MCP tool, and two rules that disagree. Why it matters: Practical help for developers...

The thread is aimed at developers whose agent instructions get ignored in long sessions, and it offers a practical list of failure modes to check. Teams relying on coding agents could use these five categories to diagnose why rules slip mid-session. Watch whether the full fixes work in real projects, since this is one source's guidance.

Teams using AI at work may want to compare this against current productivity and review workflows.

Signal Strength 81% Technical label WATCH Public Interest 29 Category AI AT WORK Reader Depth TECHNICAL

Signal Strength reflects source quality, relevance, freshness and evidence. Public Interest helps organize discovery; it is not proof of truth.

Public Interest components
Recognizable Entity Score 0 Practical Impact Score 28 Novelty Interest Score 70 Consequence Score 22 Curiosity Score 16 Shareability Score 46

VQV surfaced this signal because it is recent, relevant to AI Coding Tools, connected to Scobleizer News Pulse (X).