Report 01M7FPACFQ5Y6N2XPREKS38V3T · checks-v2
github-ops skill: automated checks
What HostedMCP checked on 15 Nov 2026, 13:28 UTC. Every result links to the public source it came from, so you can check it yourself.
7 of 7 checks ran · 0 high findings · not a security audit
- Checked
- 15 Nov 2026, 13:28 UTC
- Folder
- skills/github-ops ↗
- Files read
- SKILL.md and scripts
- Methodology
- checks-v2
What to fix for a higher score
One check can earn more points. The next scan runs within a week, or right after the repository changes.
- Pre-approved tools are scoped up to +10 pts
Narrow
allowed-toolsto the commands the skill needs, for exampleBash(git status:*), notBashorBash(curl:*). The agent still asks before anything else.
Origin
25 / 25 points- PassSource repository linked and reachable10 pts
0xdarkmatter/claude-mods/skills/github-ops is public and reachable
github.com/0xdarkmatter/claude-mods/tree/HEAD/skills/github-ops ↗ - Pass
- PassCommit or release in the last year5 pts
Last commit 2026-11-07
github.com/0xdarkmatter/claude-mods/commits ↗
Scripts
30 / 30 points- PassNo download-and-run or credential theft in scripts30 pts
3 scripts read; no download-and-run, remote eval, base64 execution or credential upload
Instructions to the agent
20 / 30 points- PassNo hidden instructions in SKILL.md20 pts
No override phrases, safeguard switches, credential requests or hidden characters
github.com/0xdarkmatter/claude-mods/blob/HEAD/skills/github-ops/SKILL. ↗ - FindingPre-approved tools are scoped10 pts
Bash lets the agent run any shell command without asking
github.com/0xdarkmatter/claude-mods/blob/HEAD/skills/github-ops/SKILL. ↗How to fix: Narrow
allowed-toolsto the commands the skill needs, for exampleBash(git status:*), notBashorBash(curl:*). The agent still asks before anything else.
Format
15 / 15 points- PassSKILL.md follows the Agent Skills format15 pts
Valid name and description; says when to use it; 264 lines
github.com/0xdarkmatter/claude-mods/blob/HEAD/skills/github-ops/SKILL. ↗
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 the skill folder 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.