Scan Report
0 /100
harness-generate-iOS
Auto-generate the full Claude Code harness for an iOS project (CLAUDE.md, docs/, README quick-cards, .claude/rules/)
The harness-generate-iOS skill is a pure documentation-generation tool with no executable code, shell access, credential harvesting, or data exfiltration capabilities.
Safe to install
This skill is safe to use. No security concerns identified.
| Resource | Declared | Inferred | Status | Evidence |
|---|---|---|---|---|
| Filesystem | NONE | READ | ✓ Aligned | Uses Read/Glob to scan project files only — no Write required in the skill itsel… |
| Network | NONE | NONE | — | No network requests declared or inferred |
| Shell | NONE | READ | ✓ Aligned | Bash usage in scan-procedures.md is read-only (xcodebuild -list, ls, grep); no c… |
| Environment | NONE | NONE | — | No environment variable access detected |
| Skill Invoke | NONE | NONE | — | Agent sub-tasks used only for parallelism, not cross-skill invocation |
| Clipboard | NONE | NONE | — | Not used |
| Browser | NONE | NONE | — | Not used |
| Database | NONE | NONE | — | Not used |
File Tree
3 files · 22.0 KB · 704 lines Markdown 3f · 704L
├─
▾
references
│ ├─
scan-procedures.md
Markdown
│ └─
templates.md
Markdown
└─
SKILL.md
Markdown
Security Positives
✓ Entirely Markdown-based skill with no executable code
✓ All tool usage is read-only (Glob, Grep, Read) for project analysis
✓ No scripts, binaries, or dependencies present
✓ No credential, key, or token access
✓ No network egress or C2 indicators
✓ No obfuscation, encoding, or suspicious patterns
✓ Skill outputs are local file writes strictly scoped to the target project (CLAUDE.md, docs/, .claude/rules/)
✓ Template output is reviewed by the user before acceptance (AUTO-GENERATED markers)