solsentry

Interpreting Scores — Development skill for Claude Code

Development community

Reference: interpreting-scores (read this before you quote a number).

How to install Interpreting Scores

Installs to ~/.claude/skills/solsentry-solana-counterparty-gate-interpreting-scores/SKILL.md

Terminal

mkdir -p ~/.claude/skills/solsentry-solana-counterparty-gate-interpreting-scores && curl -fsSL https://raw.githubusercontent.com/solsentry/solana-counterparty-gate/HEAD/skill/interpreting-scores.md -o ~/.claude/skills/solsentry-solana-counterparty-gate-interpreting-scores/SKILL.md

Restart Claude Code, or start a new session, for it to be picked up.

What Interpreting Scores does

Reference: interpreting-scores (read this before you quote a number)

This reference exists so the agent reads SolSentry output **honestly** — not over-claiming, not under-warning. Treating a score correctly is a feature, not a footnote.

The tiers (and what each actually licenses you to say)

Level	Operator criteria	Token criteria	What you may say
`CRITICAL`	10+ confirmed rugs	token confirmed as rug	"do not compose / do not sign"
`HIGH`	5+ confirme

Alternatives in Development

Sentiment Analysis — Analyze user feedback with sentiment scores and JTBD insights 7.8k ★
Quick Eval — Quick job evaluation 473 ★
Migrations — Versioned breaking change migrations for Turbo 263 ★