Scan Report
0 /100
hive-builder
Architecture and scaffolding skill for building Hive-style multi-agent systems in OpenClaw
Pure documentation/scaffolding skill with no executable code, network access, or credential handling. Well-documented safety boundaries and explicit disclaimers about what this skill is and isn't.
Safe to install
This skill is safe to use. No action required.
| Resource | Declared | Inferred | Status | Evidence |
|---|---|---|---|---|
| Filesystem | READ | READ | ✓ Aligned | SKILL.md: Only reads existing files to understand workspace state; creates new f… |
| Network | NONE | NONE | — | No network requests in any files |
| Shell | NONE | NONE | — | No shell scripts or subprocess calls in any files |
| Environment | NONE | NONE | — | No environment variable access |
| Skill Invoke | NONE | NONE | — | Mentions other hive-* skills but only as documentation, not actual invocation |
File Tree
6 files · 24.4 KB · 748 lines Markdown 5f · 743L
JSON 1f · 5L
├─
▾
templates
│ ├─
00-base-roles.md
Markdown
│ └─
README.md
Markdown
├─
_meta.json
JSON
├─
HIVE-合规专员-扩员.md
Markdown
├─
README.md
Markdown
└─
SKILL.md
Markdown
Security Positives
✓ Pure markdown/documentation skill with no executable code
✓ Comprehensive SKILL.md with clear safety boundaries and disclaimers
✓ Explicitly separates installation from activation
✓ Safe adoption rules clearly documented for existing workspace mode
✓ No shell scripts, Python, or JavaScript files present
✓ No network requests or external dependencies
✓ Mentions other hive-* skills as documentation only, not as runtime dependencies
✓ No credential harvesting, key exfiltration, or sensitive path access
✓ Templates are clearly labeled as planning materials, not deployment scripts