Item detail
github.com

kellyvv/PhoneClaw

RepoRadar surfaced kellyvv/PhoneClaw — a mobile agent — into the Local AI section, where it sits at Gold tier with a 'worth watch' verdict. Its strongest signal is novelty, scored 9.0 out of 10.

Score8.1
Popularity1.0
Riskmedium
TierGold
Score breakdown
Usefulness8.0
Novelty9.0
Momentum7.0
Maturity6.4
Open-source/build8.4
Evidence7.2
Workflow potential8.5
Setup ease4.2

Popularity is tracked separately. Support, ads, sponsorships, and tips never affect these signals.

Why it matters

Useful for mobile-agent builders who want a real iOS-native testbed for on-device models, native permissions, and phone-first agent UX.

Who should use it

Builders prototyping mobile-native AI agents on iPhoneDevelopers exploring on-device models plus native permissioned skillsResearchers studying phone-first agent UX and local memory constraintsTeams evaluating where a mobile agent should stay local versus hand off heavier inference to nearby hardware

Who should skip it

Avoid running kellyvv/PhoneClaw in production until you have reviewed its permissions, data-access scope, and failure modes in a sandbox.

About this signal

kellyvv/PhoneClaw is tracked by RepoRadar as a mobile agent in the Local AI section. It was first seen on 2026-06-30 and last updated on 2026-06-30. The current verdict is 'worth watch' with a Gold tier and advanced setup difficulty. kellyvv/PhoneClaw leads on novelty (9.0) and workflow potential (8.5); its lowest signal is setup ease (4.2), so factor that in before investing setup time. This page summarizes the evidence RepoRadar has captured from captured source metadata. The score, tier, risk label, and verdict on this page are never influenced by sponsorship, ads, or tips — they reflect only the usefulness, popularity, novelty, momentum, maturity, and evidence signals described in the RepoRadar methodology.

How this item is evaluated

RepoRadar assigned kellyvv/PhoneClaw a composite score of 8.1 out of 10, placing it in the Gold tier. This score combines weighted sub-signals: usefulness (35%), novelty (18%), momentum (14%), maturity (10%), open-source/build quality (7%), evidence quality (6%), workflow potential (6%), and setup ease (4%). Popularity is tracked separately at 1.0 and never affects the composite score or tier. The risk label of 'medium' reflects inherent user-impacting hazards, not generic novelty. Items with no risk flag may still require normal code review before production use.

Putting this into practice? Read How to vet an AI agent or MCP server before you wire it in for the checklist behind this score.

Risk explanation

Built-in Skills can read or act on local calendar, reminders, contacts, clipboard, and health data, so first evaluation belongs on a non-primary phone with only the needed permissions enabled; Optional Mac remote inference sends requests to a paired LAN machine and can route onward to Ollama, Codex CLI, or other providers, so verify that data flow before using it with sensitive personal context.

Evidence links
Closest alternatives / related signals
iphonelocal-aimobile-agentioson-deviceapache-2.0