Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
Review Ios
v1.0.0Comprehensive iOS/SwiftUI code review with optional parallel agents
⭐ 0· 59·1 current·1 all-time
byKevin Anderson@anderskev
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
The declared purpose (iOS/SwiftUI code review) matches the SKILL.md steps (git diff, SwiftLint, grep for framework usage, targeted review flow). However the skill's metadata lists no required binaries or environment variables while the instructions clearly rely on git, grep and SwiftLint — this is an inconsistency (the skill will fail or behave unexpectedly unless those tools exist).
Instruction Scope
Instructions run shell commands that read the repository (git diff, grep) and invoke SwiftLint and also tell the agent to load a verification skill and many beagle-ios skills. Loading and invoking other skills/subagents expands the effective scope and trust boundary: those other skills could request credentials, perform network calls, or run additional commands. The SKILL.md itself does not declare or justify loading these external skills, nor enumerate what those skills require.
Install Mechanism
There is no install spec and no code files — this is instruction-only, which reduces install-time risk (nothing is downloaded/extracted by the skill).
Credentials
No environment variables, credentials, or config paths are declared and the review does not itself require secrets. However, because the workflow loads external skills (beagle-ios:*), those other skills may request credentials or environment access. Also the SKILL.md fails to declare the locally required binaries (git, grep, swiftlint).
Persistence & Privilege
The skill does not request persistent/always-on privileges (always:false). It sets disable-model-invocation:true in the metadata, which is unusual given the SKILL.md instructs spawning subagents and using Skill/Task tools; this is an operational inconsistency but not itself an elevation of privilege.
What to consider before installing
This skill's content looks like a reasonable iOS code-review workflow, but it has important inconsistencies you should resolve before trusting it: (1) The runtime assumes local tools (git, grep, SwiftLint) but the metadata lists none — ensure those binaries are available or update the skill metadata. (2) The skill instructs the agent to load many external beagle-ios skills and to spawn subagents; get the metadata and trust decisions for those skills (what they install, what env vars or network access they need) before enabling this skill. (3) Because it runs shell commands against your repo, run it in a safe environment (local checkout or sandbox) and review exactly which skills will be loaded. If you need higher assurance, ask the author for the beagle-ios skill manifests, or run the workflow manually once to verify behavior.Like a lobster shell, security has layers — review code before you run it.
latestvk976rcb5rd0dvmcbngx5ahpy6584jkz0
License
MIT-0
Free to use, modify, and redistribute. No attribution required.