blindspot

Solid

Find the unknowns before they find the work — a read-only territory pass that surfaces the questions nobody knew to ask, judgment taught before a choice is extracted, and a quiz that catches the user's map up with what actually changed. Use when work enters territory the user calls unfamiliar, when the user is confirming a choice they can't evaluate, or when the user asks what they're missing or wants to be quizzed before accepting built work.

AI & Automation 5 stars 0 forks Updated 4 days ago MIT

Install

View on GitHub

Quality Score: 80/100

Stars 20%
26
Recency 20%
100
Frontmatter 20%
70
Documentation 15%
100
Issue Health 10%
80
License 10%
100
Description 5%
100

Skill Content

# Blindspot The work file, the prompt, the glossary — all map. The codebase, its history, its real constraints — territory. The gap between them is the unknowns, and an unknown left unfound doesn't stay unfound: the diff review finds it, or production does. Sort any unknown into one of four kinds; the flow already mines three: | Kind | What it is | Who finds it | |---|---|---| | Known knowns | Already written down | The work file carries them | | Known unknowns | Questions you know are open | Wonder settles or parks them | | Unknown knowns | Obvious to the user, never said | The interview extracts them | | Unknown unknowns | On nobody's question list | **Only the territory can surface them — this skill** | An interview can't reach the fourth kind: it only asks questions someone thought to ask. This skill owns three moves for the moments the map is furthest from the territory. Each feeds a stage; none replaces one. ## Move 1 — the territory pass (before the work) When work enters an area the user names as unfamiliar, when interview answers keep coming back "I don't know", or when a whole discipline nobody on the work has judgment in walks in (a new subsystem, a new domain entirely) — walk the territory before anyone writes the brief. Read only — the code, its tests, its docs, and its history (log and blame of the area; reverted commits and bug-fix clusters are where it bit last time). Spawn the pass as a fresh subagent **on a frontier-tier model — the session's main mod...

Details

Author
donald-ada
Repository
donald-ada/workinggenius
Created
3 weeks ago
Last Updated
4 days ago
Language
Shell
License
MIT

Similar Skills

Semantically similar based on skill content — not just same category

AI & Automation Listed

blindspot

Surface the USER's unknown-unknowns before they work in unfamiliar territory — an unfamiliar codebase area OR an unfamiliar domain vocabulary — and coach a sharper prompt. Scans for gaps the user's framing missed, emits one blindspot card per gap (the gap, why it matters here, a copyable prompt-fix line), then assembles the fixes into one improved implementation prompt. Use when about to work somewhere you don't know well and the goal is a better prompt, not the codebase handoff artifact /discovery:explore produces.

4 Updated today
melodic-software
AI & Automation Listed

surveying-blind-spots

Use when the user is about to start work in a codebase area or field they do not know well and wants a blind-spot pass, a survey of their unknown unknowns, gotchas, and dead ends, before they write a prompt or plan. For example "I know nothing about the auth module, find my blind spots" or "what don't I know about color grading". Not for tutoring to mastery; that is quizzing-a-topic.

0 Updated 1 weeks ago
pgoell
AI & Automation Listed

finding-unknowns

Guide the user through a quadrant walk that maps the unknowns of a task — open by listing the known knowns, then work through known unknowns, unknown knowns, and unknown unknowns one stage at a time, ending with a complete four-quadrant map in the user's hands. Use when the user explicitly asks for a blindspot pass, unknown unknowns, "what am I missing / what should I be asking", a one-question-at-a-time interview, or several directions to react to; when ambiguity or unfamiliarity is high enough that building now would likely cause rework; when a reference implementation must be understood before porting; or for a named after-walk slice — implementation notes / deviations log, a buy-in doc, a quiz gate before merge, or a session handoff (run only that slice). Do not invoke for ordinary implementation of a sufficiently specified task.

2 Updated today
avivsinai