grounded-answers

grounded-answers is a skill for Claude Code from jx-hxxx/hi-vibe. It costs 163 tokens per session (1,914 once invoked), scanned A, original, MIT.

A guidance skill for making decisions and stating facts carefully. It addresses missing requirements, ambiguous choices, and claims about numbers, limits, software behavior, or current services.

In plain words
What is it for?
Use it when planning changes, checking technical behavior, discussing limits or prices, or handling work that could be difficult to undo.
Why use it?
It reduces unsupported assumptions by requiring direct evidence or clearly labeled estimates, and prompts for clarification when a choice could materially change the result.

Skill for Claude Code

Written for Claude Code: user-invocable in frontmatter. Also seen: names the AskUserQuestion tool.

Part of the hi-vibe plugin — 6 skills, 10 commands, 2 agents, 5 hooks shipped together

Good fit Use it when planning changes, checking technical behavior, discussing limits or prices, or handling work that could be difficult to undo.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/jx-hxxx/hi-vibe/grounded-answers
Install

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.

Any agent
npx skills add jx-hxxx/hi-vibe --skill grounded-answers
Clone the repo
git clone --depth 1 https://github.com/jx-hxxx/hi-vibe

Made for: Claude Code.

Or install hi-vibe, the plugin that ships this one along with the rest of its 6 skills, 10 commands, 2 agents, 5 hooks.

Wrote 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.

agentmods badge for grounded-answers

README.md
[![agentmods](https://agentmods.dev/badge/skills/jx-hxxx/hi-vibe/grounded-answers/github.svg)](https://agentmods.dev/skills/jx-hxxx/hi-vibe/grounded-answers)
Your own site
<a href="https://agentmods.dev/skills/jx-hxxx/hi-vibe/grounded-answers"><img src="https://agentmods.dev/badge/skills/jx-hxxx/hi-vibe/grounded-answers/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.

agentmods 80×15 button for grounded-answers

Your own site · 80×15
<a href="https://agentmods.dev/skills/jx-hxxx/hi-vibe/grounded-answers"><img src="https://agentmods.dev/badge/skills/jx-hxxx/hi-vibe/grounded-answers.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 163 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,914 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00163 $0.01914
Opus 5 $0.00081 $0.00957
Sonnet 5 $0.00033 $0.00383
Haiku 4.5 $0.00016 $0.00191

Measured 12d ago against content hash 3021ee8d6384, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

grounded-answers 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 12d 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.

skills/grounded-answers/SKILL.md · 111 lines

How it starts

The opening of the file, as written. The whole thing — 111 lines — stays where its author put it; the contents beside it link to each section on GitHub.

grounded-answers

답변 언어: 이 지침은 한국어로 쓰였지만, 출력은 항상 사용자가 대화에서 쓰는 언어를 따른다 (한국어→한국어, 영어→영어). 기존 문서·코드에 언어가 있으면 그쪽을 우선한다.

두 가지 실패를 막는다: ① 헷갈리는 걸 묻지 않고 임의로 진행하는 것, ② 추정을 공식 수치처럼 말하는 것.

Part 1 — 언제 묻고, 언제 그냥 진행하는가

반드시 물어라 (AskUserQuestion):

  • 해석이 여러 개고, 어느 쪽이냐에 따라 사용자가 보는 결과가 달라질 때
  • 데이터 모델/API/화면 흐름을 좌우하는 요구사항이 빠져 있을 때
  • 되돌리기 어렵거나 파괴적인 작업 (삭제, 덮어쓰기, 마이그레이션, force push, 기존 설정 변경)
  • 의존성 추가, 비용 발생, 외부 계정/서비스 연결

묻지 말고 진행하되, 답변에 "이렇게 정했다"를 밝혀라:

  • 저장소의 기존 관례를 따르는 경우 (이름 짓기, 파일 위치, 포맷)
  • 결과가 동등한 내부 구현 세부사항
  • 업계 표준이 존재하는 경우 — 표준을 택하고 근거를 한 줄로 밝힌다
  • 파일 하나에 국한되고 즉시 되돌릴 수 있는 선택

판단 기준 한 줄: 되돌리기 어렵거나, 사용자 눈에 보이거나, 요구사항이 갈리면 → 묻는다. 아니면 관례를 따르고 선택을 밝힌다.

질문은 한 번에 모아서 (3~4개까지 한 AskUserQuestion에). 하나씩 찔끔찔끔 묻는 것도, 모든 것을 묻는 것도 실패다.

Part 2 — 수치·사실 주장의 근거

다음을 말하려면 둘 중 하나가 반드시 있어야 한다 — 숫자·제한·가격, 라이브러리/프레임워크/API 동작·필드명·파라미터, 버전 호환성, 그리고 외부 플랫폼·서비스의 현재 동작·정책·제약(예: GitHub·npm·클라우드 콘솔이 "왜 이렇게 동작하나", "이 기능 지원되나", "최근 정책이 바뀌었나"):

  1. 실제 근거 제시: 공식 문서 URL, 코드 file:line, 방금 실행한 명령의 실측 출력
  2. 추정 라벨링: "추정입니다 — 근거: ○○" 를 명시

기억(훈련 데이터)으로 답하지 마라. 훈련 데이터는 낡는다 — 특히 외부 플랫폼의 정책·제한·UI 동작은 조용히 바뀐다. 근거를 구하는 순서:

  1. 직접 실행/호출 — 명령·API를 돌려 실측 출력을 본다
  2. context7 MCP (붙어 있으면) — 라이브러리·프레임워크·API·플랫폼 공식 문서를 질의한다. 이런 사실은 여기부터 확인한다.
  3. 웹 검색 / 공식 문서 fetch — context7에 없으면 공식 문서를 직접 연다
  4. 위가 다 안 되면 → "추정입니다 — 근거: ○○" 라고 밝히고 말한다

셋 다 건너뛰고 기억으로 단정하면 안 된다. "이건 원래 이렇게 동작해요", "이 기능은 지원돼요/안 돼요"도 전부 근거가 필요한 사실 주장이다 — 대화가 코딩이 아니라 잡담·트러블슈팅이어도 예외 없다. 확인 도구가 있는데(context7 등) 안 쓰고 추측하는 것이 가장 흔한 실패다.

이 저장소에 대한 구조적 주장(중복, 미사용, 존재 여부)은 repo-xray 스킬의 계약을 따른다 — 스캔 없이 개수를 말하지 않고, 스캔 범위 없이 "없다"고 말하지 않는다. 같은 원칙을 외부 사실로 확장한 것이 이 스킬이다.

Part 3 — 진단·상태 판단도 근거가 필요하다

"느리다 = 고장"이 아니다. "멈췄다 / 무한이다 / 안 돌아간다 / 이거 때문에 안 된다"는 전부 근거가 필요한 진단 주장이다 — 수치 주장과 똑같이 다뤄라. 느린 것과 멈춘 것은 다르고, 이 둘을 구별하는 유일한 방법은 **방해 없이 한 번 끝까지 재보는 것(1회 실측)**이다.

확인 전에 파괴적 재시도를 하지 마라. 오래 걸리는 명령을 타임아웃으로 끊고 → 다시 돌리고 → 또 끊는 것, 죽었는지 확인 안 하고 프로세스를 여러 개 겹쳐 띄우는 것은 진단 대상을 오염시키고(부하가 늘어 더 느려짐) 원인을 영영 못 찾게 만든다. 순서는 반대다:

Read the full file on GitHub · 111 lines

Changes

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.

  1. 12d ago First seen · 111 lines · 163 tokens per session scan A 3021ee8d6384

Subscribe to this mod's changes

grounded-answers is a skill published in the GitHub repository jx-hxxx/hi-vibe (3 stars, last pushed today), licensed MIT. It adds 163 tokens to every session and 1,914 once invoked, about $0.0008 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 2026-08-31.

Related

Other skills, from other repositories

procoder

Work like a senior developer in a repository governed by procoder: run the commit gate before calling anything done, format and lint through the binary, and drive the spec, plan, todo, backlog, and sprint chain in .procoder/. Use this skill when the repository contains a .procoder/ directory or an AGENTS.md naming…

azrtydxb/procoder · 101 tokens

code-review

The depth half of a review - the dimensions a diff is read against (correctness, boundaries, concurrency, failure paths, secrets, data access, structure, test quality) and the rule that a finding is refuted before it is reported. The verdict stays with the reviewer agent. Use when reviewing a diff or a pull request…

jjanczur/tyran · 80 tokens

fidelity-gate

Build a UI against a frozen visual reference without drift - an inventory extracted before any code, a relics list, and a gate that MEASURES computed styles on a fixture carrying the reference's own data. Use when a mockup, design spec or screenshot is the contract.

jjanczur/tyran · 56 tokens

deslop

The optimization pass, defined - delete before you add, one smell class per pass, behaviour pinned by a test that ran BEFORE the edit. Lints a SKILL.md and prose by the same instinct. Use for the per-story optimization pass or when code has grown noisy without growing capable.

jjanczur/tyran · 58 tokens

prompt-tuning

Tune a prompt, or anything whose quality is measured by non-deterministic model output, without chasing noise - a noise baseline before the first edit, medians over repeated runs, enforcement AFTER generation rather than in the wording. Use when iterating on prompts or model-judged output.

jjanczur/tyran · 58 tokens

root-cause

Find the mechanism behind a failure instead of patching its symptom - reproduce first, one variable per experiment with the prediction written before the run, exit by naming the mechanism and pinning it with a failing test. Use for a bug, an unexplained red test, or a failure that will not reproduce.

jjanczur/tyran · 61 tokens