Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/ferrumclaudepilgrim/claude-code-android/termux-safenpx skills add ferrumclaudepilgrim/claude-code-android --skill termux-safegit clone --depth 1 https://github.com/ferrumclaudepilgrim/claude-code-androidWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/ferrumclaudepilgrim/claude-code-android/termux-safe)<a href="https://agentmods.dev/skills/ferrumclaudepilgrim/claude-code-android/termux-safe"><img src="https://agentmods.dev/badge/skills/ferrumclaudepilgrim/claude-code-android/termux-safe.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00036 | $0.00916 |
| Opus 5 | $0.00018 | $0.00458 |
| Sonnet 5 | $0.00007 | $0.00183 |
| Haiku 4.5 | $0.00004 | $0.00092 |
Grade A, and why
termux-safe scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 47 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Android / Termux Environment Constraints
Scope and when this applies. These constraints apply when you are executing inside Termux on Android, regardless of how Claude Code itself was packaged. The load-bearing runtime signal is
$PREFIX === "/data/data/com.termux/files/usr". If that environment variable points at Termux's prefix, you are in Termux and these constraints apply, even ifprocess.platformreports"linux"(it will report"linux"if Claude Code was installed via this repo's v2.9.0 Path A, which runs the linux-arm64 binary patched via glibc-runner; it will report"android"only on older Path A v2.x installs that used the JS bundle). If$PREFIXis unset or points at/usr, you are inside a chroot-like sandbox (for example, a Linux rootfs running underproot-distro, Path B) or inside a hypervisor-managed Linux VM (Path C), and standard Linux conventions apply; most of this file does not.
If the runtime signals indicate the in-Termux case, the constraints below produce silent failures, not errors. Every suggestion must account for them.
Hard Rules: Never Suggest These
- No
sudo. Root does not exist. Nosu, nodoas, no privilege escalation of any kind. - No
systemctlorsystemd. Android does not use systemd. Nojournalctl, noservice, no unit files. - No standard Linux paths.
/usr/bin,/etc,/vardo not exist or are not writable. Termux paths:- Home:
/data/data/com.termux/files/home - Prefix:
/data/data/com.termux/files/usr - Binaries:
$PREFIX/bin - Config:
$PREFIX/etc
- Home:
- No ports below 1024. No root = no binding to 80, 443, etc. Use 1024+.
- Prefer
pkgoverapt. Both work, butpkgauto-updates the package index before installing. - No Docker, no containers. The kernel does not support them without root.
Silent Failure Modes
- Node.js is not required by this repo's
install.sh. The patched linux-arm64 claude binary is self-contained. If Node is installed separately on native Termux, prefer v25+: v24 had a documented startup hang on ARM64 that was resolved in v25. - File descriptor limits vary by device. Check with
ulimit -n. Avoid spawning many concurrent processes. - Android phantom process killer limits background processes to ~32 across all apps. If "Disable child process restrictions" is enabled in Developer Options, this limit is lifted; 6 concurrent subagents have been observed stable in testing. Otherwise limit to 2-3.
process.platformreporting varies by install path. On Path A v2.9.0 (the linux-arm64 binary patched via glibc-runner), Node.js reportsprocess.platform === "linux"even though you are inside Termux. On older Path A v2.x (the JS-bundle npm install) it reports"android". On Path B (proot-distro Ubuntu) and Path C (AVF), it reports"linux"and you are not in Termux. The reliable cross-path signal that you are still in Termux is$PREFIX === "/data/data/com.termux/files/usr".
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 5d ago First seen · 47 lines · 36 tokens per session scan A 356c8306c77c
termux-safe is a skill published in the GitHub repository ferrumclaudepilgrim/claude-code-android (222 stars, last pushed 1mo ago), licensed MIT. It adds 36 tokens to every session and 916 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
analyzing-android-malware-with-apktool
Perform static analysis of Android APK malware using apktool for resource decompilation, jadx for Java source recovery, and androguard for manifest inspection, dangerous permission-combination detection, and identification of obfuscated code, dynamic code loading, and reflection-based API calls. Use to statically…
solopi-ai
通过 SoloPi 的机器可读 CLI 编译和执行 AI 验证计划，管理签名端侧 ExecuTorch 决策模型、持久设备池、无人值守任务、安卓设备、应用、动作、配置、用例步骤与交互录制、回放及性能历史、动态 Agent、批量与重复执行、性能监控、压力测试和证据。适用于需求/AC 到 Result Judge 三态结论、cloud/on-device 决策切换、模型发布门禁，以及 generation 租约的多设备 CI 执行。.
argent-tv-interact
Control and inspect TV apps via argent — Apple TV (tvOS), Android TV (leanback), and Amazon Fire TV (Vega). Boot the target, read focus, navigate with the D-pad remote, type, screenshot, and on Vega debug the JS runtime (evaluate, console logs, network inspector). Use when a task targets a TV (runtimeKind "tv", or…
compose-recomposition-performance
Use when investigating Jetpack Compose recomposition performance, skippable/restartable composables, composables.txt or compiler reports, Layout Inspector recomposition counts, or frame-rate State reads in composition vs layout/draw, and it is not yet clear whether the cause is parameter stability or deferred reads.…
android-navigation-3
Install and migrate to Jetpack Navigation 3. Use when implementing Navigation 3 patterns including NavDisplay, NavKey routes, deep links, multiple backstacks, scenes (dialogs, bottom sheets), or migrating from Navigation 2.
ios-slim-bindings
Create iOS slim bindings for MAUI. USE FOR: slim iOS binding, Native Library Interop, Swift/Objective-C wrappers, XcodeGen project.yml, Podfile, CocoaPods static linking, BUILDLIBRARYFORDISTRIBUTION, XcodeProject MSBuild, @objc/[Export] selector crashes, async completion handlers. DO NOT USE FOR: Android bindings…