graphql-schema

[Skill Scanner] [Documentation context] Backtick command substitution detected This skill implements reasonable GraphQL schema discovery, validation, federation checks, and reporting consistent with its stated purpose. I found no direct malicious behavior (no exfiltration, no backdoor, no obfuscated payload). The primary security concern is supply-chain risk from unpinned npm installs and runtime generation/execution of validate-schema.js — these increase the chance that an attacker compromising the npm packages (or if a package name is altered) could run malicious code in the project environment. Recommend pinning dependency versions, using package lockfiles, and auditing installed packages before execution; optionally run validation in an isolated environment (container or CI runner) rather than locally with direct npm installs. LLM verification: [LLM Escalated] This skill is functionally consistent with its stated purpose (GraphQL schema validation and optimization). It does not contain obvious malicious code (no network exfiltration, no backdoor/reverse shell, no hardcoded credentials). However, there are supply-chain and operational risks: it runs unpinned npm installs (modifies the project environment and pulls code from npm without version pins), and it writes and executes a generated JS file. These patterns increase the attack surface and make the