Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add sangrokjung/claude-forge --skill blind-spot-passgit clone --depth 1 https://github.com/sangrokjung/claude-forgeWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/sangrokjung/claude-forge/blind-spot-pass)<a href="https://agentmods.dev/skills/sangrokjung/claude-forge/blind-spot-pass"><img src="https://agentmods.dev/badge/skills/sangrokjung/claude-forge/blind-spot-pass/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/sangrokjung/claude-forge/blind-spot-pass"><img src="https://agentmods.dev/badge/skills/sangrokjung/claude-forge/blind-spot-pass.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00221 | $0.02670 |
| Opus 5 | $0.00111 | $0.01335 |
| Sonnet 5 | $0.00044 | $0.00534 |
| Haiku 4.5 | $0.00022 | $0.00267 |
Grade A, and why
blind-spot-pass scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 137 lines — stays where its author put it; the contents beside it link to each section on GitHub.
blind-spot-pass — surface your unknown unknowns before you start
(착수 전 나의 맹점 발굴 + 사전 학습)
Analogy. Before asking for directions in an unfamiliar city, a good local guide first tells you "what to watch out for here, what's worth seeing, the words people use." Only then do you know what to ask. This skill gives you that "local briefing" right before you hand off work in a domain you don't know well.
비유. 낯선 도시에서 길을 묻기 전에, 현지 가이드가 먼저 "이 동네에서 조심할 것, 꼭 봐야 할 것, 여기 사람들이 쓰는 말"을 짚어준다. 그러면 이제 무엇을 물어야 할지 스스로 알게 된다. 이 스킬은 낯선 분야의 작업을 맡기기 직전, 그 "현지 브리핑"을 해주는 역할이다.
Why it matters (왜 필요한가)
Since Fable-class models, the bottleneck on output quality is no longer model capability — it's how well you surface what you don't know. When you hand off work in an unfamiliar domain, you don't even know what to ask (unknown unknowns). Prompt too vaguely and the model silently applies some industry default; prompt too specifically and it follows your wrong turn off a cliff. This skill drains that blind spot before real work starts, because discovering it mid-build is expensive to undo (cost asymmetry / shift-left).
한글: Fable 이후 결과물 품질의 병목은 모델 성능이 아니라 "내가 무엇을 모르는지"를 얼마나 짚어내느냐다. 낯선 분야를 맡길 때는 무엇을 물어야 할지조차 모르는 상태(unknown unknowns)라, 너무 막연하면 모델이 업계 표준을 임의로 갖다 쓰고, 너무 구체적이면 틀린 방향도 고집한다. 이 스킬은 착수 전에 사각지대를 값싸게 드러낸다 — 구현 중에 발견하면 되돌리기 비싸기 때문이다.
The real deliverable is not a "list of questions" — it's a method that surfaces blind spots and teaches the minimum. If it degrades into "let me ask you some questions", it just duplicates planner. Don't let it. (진짜 산출물은 질문 목록이 아니라 사각지대를 드러내고 최소한만 가르치는 방법이다.)
The three kinds of unknowns (세 가지 불확실성)
| Kind | What it is | Handled by |
|---|---|---|
| Known knowns | what you wrote in the prompt (프롬프트에 적은 것) | just instruct |
| Known unknowns | undecided, but you know you don't know (모른다는 건 아는 미결정) | planner interview / /plan |
| Unknown knowns | so obvious you didn't say it — tacit taste, "I know it when I see it" (너무 당연해 안 적은 암묵지) | /explore references · lightweight prototype |
| Unknown unknowns | never even considered — an unfamiliar domain (아예 고려조차 못 한 것) | this skill |
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 11d ago First seen · 137 lines · 221 tokens per session scan A d92a49348f76
blind-spot-pass is a skill published in the GitHub repository sangrokjung/claude-forge (835 stars, last pushed 8d ago), licensed MIT. It adds 221 tokens to every session and 2,670 once invoked, about $0.0011 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
career-path-planner
Career goal mapping with skill gap analysis, actionable development plans, and milestone tracking. Use when planning career transitions, identifying skill gaps, setting professional development goals, or evaluating career options.
case-interview-practice
Interactive consulting case interview practice with structured frameworks, feedback mechanisms, and progressive difficulty. Use when preparing for management consulting interviews, case competitions, or business problem-solving exercises.
course-material-creator
Create custom educational materials including syllabi, lesson plans, assessments, reading lists, and learning objectives. Use when building course curricula, training programs, workshops, or educational content.
debate-practice-coach
Interactive debate and argument practice with structured feedback, counterargument generation, logical fallacy identification, and scoring rubrics. Use when preparing for debates, improving persuasion skills, or practicing argumentation.
stock-analysis-lead
Orchestrate a US-stock investment analysis — classify sector archetype, fetch SEC filings, dispatch a tiered fan-out of six vertical equity-research agents (business model, earnings quality, balance sheet, management, industry, peer comparison) over a validated JSON findings contract, then synthesize a buy/hold/sell…
kafka-event-driven-design
Kafka event-driven architecture designer and reviewer, at the application/client layer. ALWAYS use when designing, reviewing, or troubleshooting how a service produces or consumes Kafka events — topic and partition-key design, producer and consumer client configuration, consumer group topology, event schema definition…