Swift Code Review
v1.2.0Reviews Swift code for concurrency safety, error handling, memory management, and common mistakes. Use when reviewing .swift files for async/await patterns,...
⭐ 0· 105·0 current·0 all-time
byKevin Anderson@anderskev
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description (Swift concurrency, error handling, memory management) match the included README-style SKILL.md and reference docs. The skill requires no binaries, credentials, or config paths — all appropriate for a static code-review guidance skill. Note: registry metadata lacks a homepage or known source, which reduces provenance but does not contradict the skill's purpose.
Instruction Scope
SKILL.md contains a bounded checklist and points to local reference files for concurrency, error handling, observability, and common mistakes. It does not instruct reading unrelated system files, accessing environment variables, or communicating with external endpoints. All runtime instructions stay within the stated review scope.
Install Mechanism
No install spec and no code files that execute are present — instruction-only. This minimal footprint is proportionate for a documentation/checklist skill and has low installation risk.
Credentials
The skill declares no required environment variables, secrets, or config paths, which is appropriate for a static code-review helper. There are no requests for unrelated credentials.
Persistence & Privilege
always is false (normal) and disable-model-invocation is false (agent may invoke the skill autonomously). Autonomous invocation is platform-default and not problematic here, but keep in mind that the agent could call this skill without an explicit user prompt if platform policies allow.
Assessment
This skill is an instruction-only Swift code-review checklist and looks internally consistent: it needs no installs or credentials. Before installing, consider provenance (no homepage/author info) and your data-sharing policy: the skill itself doesn't exfiltrate data, but the agent's model calls may send code to the platform provider — avoid running on sensitive proprietary code unless you trust the platform's data handling. If you want stronger provenance, prefer skills with a verifiable homepage or known author and monitor future updates. If you need offline/local review only, do not enable autonomous invocation or avoid installing skills that trigger networked model calls.Like a lobster shell, security has layers — review code before you run it.
latestvk97335fa3bnm77z57d5rd66e8583b774
License
MIT-0
Free to use, modify, and redistribute. No attribution required.