Stop Your AI Agent Repeating the Same Mistake: Reviewed Skills with lessonweaver
Last Updated on August 19, 2026 by Editorial Team
Author(s): Diogo Santos
Originally published on Towards AI.
A deterministic, human-gated tool that turns your agent’s real failures into reviewed AGENTS.md, Claude, and Copilot instructions — no LLM in the loop.
Your coding agent reviewed a pull request last Tuesday. It read the title and the description, said “looks good,” and approved it — without ever opening the diff. A human caught it, left a correction, and moved on.
The article explains why agentic systems repeatedly make the same mistakes—because corrections don’t persist in durable context—and argues for a middle path between fully automated self-editing (unsafe and hard to audit) and manual updates (doesn’t scale). It introduces lessonweaver, which mines execution traces for recurring failure signals, routes candidate lessons through structured human review, gates promotion with lint checks, and exports only approved guidance as diff-first, reviewable instruction artifacts (e.g., AGENTS.md fragments, Claude skills/rules, and Copilot instructions). It then walks through the concrete five-step pipeline (detect → interview → answer → approve → export), shows how the approve stage blocks incomplete reviews unless explicitly overridden in an auditable way, and describes runtime loading via lexical retrieval with a character budget so agents start each run already equipped with the lessons a human validated. Finally, it covers limitations (early alpha status, conservative rule-based detection, trace format requirements, redaction as best-effort), positions lessonweaver within a broader “Weaver Stack,” summarizes key takeaways, and invites readers to try it on their own traces and report what the detector misses.
Read the full blog for free on Medium.
Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming a sponsor.
Published via Towards AI
Towards AI Academy
We Build Enterprise-Grade AI. We'll Teach You to Master It Too.
15 engineers. 100,000+ students. Towards AI Academy teaches what actually survives production.
Start free — no commitment:
→ 6-Day Agentic AI Engineering Email Guide — one practical lesson per day
→ Agents Architecture Cheatsheet — 3 years of architecture decisions in 6 pages
Our courses:
→ AI Engineering Certification — 90+ lessons from project selection to deployed product. The most comprehensive practical LLM course out there.
→ Agent Engineering Course — Hands on with production agent architectures, memory, routing, and eval frameworks — built from real enterprise engagements.
→ AI for Work — Understand, evaluate, and apply AI for complex work tasks.
Note: Article content contains the views of the contributing authors and not Towards AI.
Related posts
Recent Posts
