lean-mobile-audit

스타4

포크0

업데이트2026년 7월 18일 15:25

Audit mobile apps for battery and network waste on user devices — wake locks, background sync frequency, polling where push serves, location over-sampling, image/payload sizes on cellular, and retry storms on flaky networks. Use this skill whenever the user shares Android/iOS code or configs (WorkManager, background tasks, sync adapters), asks why an app drains battery or data, or wants a mobile app sustainability/efficiency review. Part of Lean Agentic AI Skills; emits lean-findings.json.

설치

Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.

출처

navveenb

navveenb/lean-agentic-ai-skills

GitHub 저장소 열기 Creator 저장소 보기

다운로드

Green Mobile Audit

Producer skill. Input: app source/configs (manifest, background task schedules, networking code), or a behavioral description. Output: lean-findings.json.

Mobile inefficiency is E shifted onto millions of user devices — and the user pays it in battery and data. It is also an R story: the same app function delivered with less device work. The physics to keep in mind: the radio is the battery's biggest enemy after the screen, and every network wakeup costs a "tail" of energized-radio time beyond the transfer itself.

Subject type: emit subject.type: "mobile-app" in findings.

Signatures

Polling where push serves — fixed-interval fetch loops for data that changes rarely; every poll wakes the radio. High. Fix: FCM/APNs push, or exponential/adaptive intervals at minimum.
Frequent background sync — WorkManager/BackgroundTasks intervals tighter than the data's real change rate; sync not batched or coalesced. High. Fix: widen intervals, use constraint-based scheduling (charging + unmetered for heavy work — the OS's own carbon-aware knob).
Wake locks held wide — partial wake locks around long code paths; alarms with setExactAndAllowWhileIdle where inexact serves. High. Fix: narrow scope, prefer WorkManager over alarms, exact alarms only for user-facing deadlines.
Location over-sampling — high-accuracy GPS at high frequency for coarse needs (geofencing, city-level). High. Fix: lowest accuracy/frequency serving the feature; passive/fused providers; geofence APIs instead of continuous tracking.
Full-size assets over cellular — images/videos not sized to device or connection class; no modern formats. Medium-high (pairs with web-carbon-audit patterns).
Retry storms — network retries without backoff/jitter on flaky connectivity; the radio-tail cost multiplies. Medium-high. Fix: exponential backoff, offline queue, respect OS connectivity callbacks instead of probing.
Chatty APIs — many small requests per screen where one aggregate serves; each carries the radio-tail overhead. Medium.
Ignoring battery saver / data saver signals — app behaves identically under OS power/data constraints. Medium; fix: honor PowerManager/DataSaver states.

Honesty rules

Battery impact is device- and usage-dependent — findings are directional; the measurement pointer is platform profilers (Android Battery Historian/Perfetto, Xcode Energy Log), named in every report. Trade-offs stated: push infrastructure has server-side cost; wider sync intervals trade freshness — name the staleness each recommendation accepts.

Cost signal (countable)

Cost driver on the provider side is minor for many findings; the honest costs of mobile inefficiency are user-side (battery, data-plan bytes). Where a provider-side call volume applies (push infra, sync endpoints), name it in cost_signal; otherwise skip the field and address the user-side cost in the impact text.

Not this skill's job

Server-side efficiency of the APIs the app calls (db-efficiency-audit, caching-strategy-designer), on-device ML placement (inference-placement-advisor).

이 저장소의 다른 Skills

같은 저장소

agentic-loop-auditor

navveenb/lean-agentic-ai-skills

Audit agent frameworks and agentic workflows for token-burning loop pathologies — unbounded retries, excessive planning depth, tool-call thrashing, context snowballing, missing loop budgets and kill switches. Use this skill whenever the user shares agent configs or traces (LangChain/LangGraph, CrewAI, AutoGen, custom loops), complains an agent is slow or expensive, or is designing any autonomous/multi-step agent. Part of Lean Agentic AI Skills; emits lean-findings.json.

2026-07-184

ai-eval-efficiency-audit

navveenb/lean-agentic-ai-skills

Audit AI evaluation infrastructure for wasted compute — full benchmark suites re-run on unchanged cases, LLM-as-judge grading without caching, oversized judge models, redundant eval passes per commit, and missing result reuse. Use this skill whenever the user shares eval harness configs or CI eval steps, complains that evals are slow or expensive, mentions LLM-as-judge costs, or runs benchmark suites on every change. Part of Lean Agentic AI Skills; emits lean-findings.json.

2026-07-184

api-efficiency-audit

navveenb/lean-agentic-ai-skills

Audit API design for chattiness and over-fetching — REST endpoints requiring N calls per screen, GraphQL resolvers with N+1 fan-out, default-everything payloads, missing pagination and field selection, provider-side polling where webhooks serve, and uncompressed responses. Use this skill whenever the user shares API schemas (OpenAPI, GraphQL SDL), resolver code, endpoint designs, or asks why clients make too many calls or payloads are heavy. Part of Lean Agentic AI Skills; emits lean-findings.json.

2026-07-184

caching-strategy-designer

navveenb/lean-agentic-ai-skills

Design caching layers so work is done once, not per request — HTTP/CDN caching, application-level caches (Redis/Memcached), computed-result caching, and cache-invalidation strategy. Use this skill whenever the user wants to add caching, asks why the same work is recomputed, designs a read-heavy system, or an audit skill (db-efficiency-audit, web-carbon-audit) flags uncached repeated work. Part of Lean Agentic AI Skills; emits a cache design plus lean-findings.json.

2026-07-184

carbon-aware-scheduler

navveenb/lean-agentic-ai-skills

Design time-shifting of flexible workloads (batch jobs, CI, training, backups, crons) to hours and regions when grid electricity is cleanest, using carbon-intensity signals like Electricity Maps, WattTime, or the Carbon Aware SDK. Use this skill whenever the user mentions carbon-aware computing, scheduling flexible/deferrable jobs, demand shifting, running workloads when energy is clean, or asks how to reduce the carbon intensity (not just amount) of their compute. Part of Lean Agentic AI Skills; emits schedule designs plus lean-findings.json.

2026-07-184

carbon-budget-designer

navveenb/lean-agentic-ai-skills

Design per-feature and per-service efficiency budgets — page-weight budgets, token budgets per AI feature, query budgets per endpoint, GPU-hour budgets per training project — with owners, review checkpoints, and enforcement hooks. Use this skill whenever the user wants sustainability targets for teams, asks how to keep efficiency from regressing organizationally, mentions performance-budget-style governance for carbon/energy, or is setting engineering standards. Part of Lean Agentic AI Skills; an advisor emitting budget definitions plus lean-findings.json.

2026-07-184

name

lean-mobile-audit

description

Green Mobile Audit

Producer skill. Input: app source/configs (manifest, background task schedules, networking code), or a behavioral description. Output: lean-findings.json.

Subject type: emit subject.type: "mobile-app" in findings.

Signatures

Polling where push serves — fixed-interval fetch loops for data that changes rarely; every poll wakes the radio. High. Fix: FCM/APNs push, or exponential/adaptive intervals at minimum.
Frequent background sync — WorkManager/BackgroundTasks intervals tighter than the data's real change rate; sync not batched or coalesced. High. Fix: widen intervals, use constraint-based scheduling (charging + unmetered for heavy work — the OS's own carbon-aware knob).
Wake locks held wide — partial wake locks around long code paths; alarms with setExactAndAllowWhileIdle where inexact serves. High. Fix: narrow scope, prefer WorkManager over alarms, exact alarms only for user-facing deadlines.
Location over-sampling — high-accuracy GPS at high frequency for coarse needs (geofencing, city-level). High. Fix: lowest accuracy/frequency serving the feature; passive/fused providers; geofence APIs instead of continuous tracking.
Full-size assets over cellular — images/videos not sized to device or connection class; no modern formats. Medium-high (pairs with web-carbon-audit patterns).
Retry storms — network retries without backoff/jitter on flaky connectivity; the radio-tail cost multiplies. Medium-high. Fix: exponential backoff, offline queue, respect OS connectivity callbacks instead of probing.
Chatty APIs — many small requests per screen where one aggregate serves; each carries the radio-tail overhead. Medium.
Ignoring battery saver / data saver signals — app behaves identically under OS power/data constraints. Medium; fix: honor PowerManager/DataSaver states.

Honesty rules

Cost signal (countable)

Not this skill's job

Server-side efficiency of the APIs the app calls (db-efficiency-audit, caching-strategy-designer), on-device ML placement (inference-placement-advisor).