Safety rating · all kinds

NVIDIA/SkillSpector

Score 20/100 — Limited public signals

Automated checks of public signals as of the date shown. Not a security audit, certification or endorsement, and not a statement about the people who maintain the project. Signals can be incomplete or out of date; each check shows its evidence and source so anyone can verify it.

Checks

• Credential and environment access — Not found (0/15): The scan flagged this check: The scanned bytes contain credential or environment access patterns: environment variable reads (process.env / os.environ / getenv); secret-named variables (API_KEY, SECRET, TOKEN, PASSWORD and similar names); credential file paths (.env, .ssh, credential stores); authorization headers and bearer values. The receipt names the pattern kinds only; no credential value is reproduced here.
• Known-bad indicators — Not found (0/15): The scan flagged this check: The scanned bytes match these known-bad indicators from the scanner's published pattern list: pipe-to-shell: a download piped straight into a shell interpreter; token- or credential-taking vocabulary (stealer, grabber, exfiltrate). Each match is a reproducible pattern match on the bytes named by the hash; re-running the published check on the same bytes reaches the same finding.
• Shell-outs and download-and-execute — Not found (0/12): The scan flagged this check: The scanned bytes contain shell-out or download patterns: child process execution (child_process, exec, spawn); shell invocation from Python (subprocess, os.system, Popen); shell command tools named in the bytes (curl, wget, powershell); download-and-execute pattern (fetched content passed to an executor in the same file). Code that runs shell commands acts with the full permissions of whoever runs it.
• Install and postinstall hooks — Found (12/12): The scan recorded a pass: No install or postinstall hooks were found in the scanned bytes.
• Obfuscation and encoded payloads — Not found (0/10): The scan flagged this check: The scanned bytes carry obfuscation or encoding signs: eval or Function-constructor evaluation of text as code. Encoded content cannot be read from the bytes alone, so what it does cannot be checked from the bytes alone either.
• Outbound network references — Not found (0/10): The scan flagged this check: The scanned bytes reference outbound network activity to these domains: api-docs.deepseek.com, api.anthropic.com, api.groq.com, api.skill.io, docs.nvidia.com, example.openai.azure.com, generativelanguage.googleapis.com, github.com, img.shields.io, modelcontextprotocol.io, my-gateway.example.com, osv.dev, platform.claude.com, www.apache.org, www.python.org. Any resource that reaches the network can send data out as well as bring data in; the findings name what the bytes reference, and the reader decides whether that fits the resource's stated purpose.
• Permission declarations vs behavior — Not found (0/10): The scan flagged this check: The bytes show network, credentials behavior, and no permissions declaration was found in the scanned bytes to compare it against.
• Provenance facts (weak signals) — Found (6/6): The scan recorded a pass: Provenance facts, labeled weak signals — they say where the bytes sit, not what the bytes do: source host: github.com; repository: NVIDIA/SkillSpector; bytes scanned: 58475; fetched from: https://raw.githubusercontent.com/NVIDIA/SkillSpector/HEAD/README.md.