메뉴
Forces Claude to extract exact quotes from files before making claims about them. Prevents hallucinations by grounding every statement in actual text from the codebase. Use when working with large files or unfamiliar code.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Complete methodology for preventing AI hallucinations. Use when accuracy is critical and you need Claude to verify before claiming, cite before asserting, and admit uncertainty instead of guessing.
Forces every code-related claim to include file:line citations. No citation means the claim must be verified or removed. Use when building documentation, writing code reviews, or any output that others will rely on.
Assign confidence scores (0-100) to every claim in a response. Helps users understand which parts are verified facts and which are educated guesses. Use when the user needs to know how much to trust each part of the answer.
Verify every claim from multiple independent angles -- code, docs, tests, git history. Use when you need high confidence that something is correct before acting on it.
Multi-angle review of responses before delivery. Audits from 4 perspectives -- factual accuracy, logical consistency, completeness, and safety. Use for high-stakes responses where errors have real consequences.
Ensure every factual claim in a response is backed by verifiable evidence from the codebase, documentation, or external sources. Use when preparing technical documentation, code reviews, or any output where accuracy is non-negotiable.
| name | context-grounding |
| description | Forces Claude to extract exact quotes from files before making claims about them. Prevents hallucinations by grounding every statement in actual text from the codebase. Use when working with large files or unfamiliar code. |
Read first. Quote second. Claim third. Never skip a step.
Before making any claim about code, follow this sequence:
Read the actual file. Don't work from memory.
Pull the exact text that supports your claim. Include line numbers.
Now state what you believe, citing the quote as evidence.
Wrong approach: "The auth middleware checks for JWT tokens and returns 401 if invalid." (No evidence. Could be hallucinated.)
Grounded approach: "Reading src/middleware/auth.ts:15-25, the middleware does:
const token = req.headers.authorization?.split(' ')[1];
if (!token) return res.status(401).json({ error: 'No token provided' });
So it extracts the Bearer token from the Authorization header and returns 401 if missing."
Always ground when:
For files over 100 lines:
For files over 500 lines:
Before sending a response, test each factual statement: "Could someone verify this by reading the file I'm referencing?"
If yes: the statement is grounded. If no: you need to add the file reference and relevant quote. If you don't have a file to reference: either find one or mark as unverified.