REPORT 01M7CJ5V59H5A28MC1ADRQ22M1 · checks-v2
ghost-scan-code skill: automated checks
What HostedMCP checked on 14 Nov 2026, 08:18 UTC in ghostsecurity/skills/plugins/ghost/skills/scan-code on the default branch. Every result links to the public source it came from, so you can check it yourself.
Passed HostedMCP checks v2
90/100
6 of 6 checks ran · 0 high findings · not a security audit
- Checked
- 14 Nov 2026, 08:18 UTC
- Files read
- SKILL.md and scripts
- Methodology
- checks-v2
Origin
35 / 35 points- PassSource repository linked and reachable15 pts
ghostsecurity/skills/plugins/ghost/skills/scan-code is public and reachable
github.com/ghostsecurity/skills/tree/HEAD/plugins/ghost/skills/scan-co ↗ - Pass
- PassCommit or release in the last year10 pts
Last commit 2026-09-28
github.com/ghostsecurity/skills/commits ↗
Scripts
30 / 30 points- PassNo download-and-run or credential theft in scripts30 pts
1 script read; no download-and-run, remote eval, base64 execution or credential upload
Instructions to the agent
25 / 35 points- PassNo hidden instructions in SKILL.md25 pts
No override phrases, safeguard switches, credential requests or hidden characters
github.com/ghostsecurity/skills/blob/HEAD/plugins/ghost/skills/scan-co ↗ - FindingPre-approved tools are scoped10 pts
Bash lets the agent run any shell command without asking
github.com/ghostsecurity/skills/blob/HEAD/plugins/ghost/skills/scan-co ↗
What these checks do not cover
- Scripts are read, never run. Their logic is not audited, and up to five scripts per skill are read.
- Results describe ghostsecurity/skills/plugins/ghost/skills/scan-code on the date above; a later commit can differ.
- A pass is not a guarantee of safety. Read the scripts before you let an agent run them, and keep allowed-tools narrow.