Skip to content
Get packs

Blind Spot Scanner

Finds one thinking blind spot in yesterday's AI chats and links a matching WeRead chapter

Source Download skill
Repository
lijigang/ljg-skills
Stars
7,607
Licence
MIT
Files
3 files

What it does

This skill reads your AI conversations from yesterday (or a date you choose), looks for a single recurring thinking habit that keeps some kind of truth out of view, and picks one book chapter from WeChat Reading (微信读书) to help close that gap. It then writes a Markdown analysis note with the blind spot, the evidence from your chats, and a web link to the chapter. It is meant for people who want a reflective, personal review rather than a quick answer.

Great for

  • People who use an AI assistant daily and want a periodic check on their own reasoning habits
  • Chinese-speaking readers who already use WeChat Reading and want targeted reading suggestions
  • Users comfortable with a personal notes folder (~/Context) that the agent writes to

Skip it if

  • Anyone who does not want their AI conversation history read and analysed; the skill scans session files under ~/.codex/sessions, which may contain sensitive personal or work content
  • Users without a WeChat Reading account or API key, since the book search and chapter steps depend on it
  • Quick factual lookups; the skill is designed for reflection and produces long analysis notes

Try asking

  • “扫盲区”
  • “看看我的思维盲区”
  • “/ljg-blind 2025-06-01 (scan a specific past date)”
  • “我昨天想漏了什么”
Step-by-step guide: how to use Blind Spot ScannerInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 3 original files, the MIT licence and an attribution note.

    ljg-blind.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip ljg-blind.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip ljg-blind.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Pass a date as YYYY-MM-DD to analyse a day other than yesterday; a day with little conversation will honestly produce no blind spot.
  • Review the scanned material path (/tmp/ljg-blind-DATE.txt) and the generated note before sharing them, since they contain your raw prompts.
  • Check that the generated WeRead link opens the correct chapter, because the skill depends on the bundled encoder rather than a documented WeRead API for links.

What's inside

  • SKILL.md
  • Tools/WeReadWebUrl.help.md
  • Tools/WeReadWebUrl.ts

Redistributed unmodified from lijigang/ljg-skills under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.