锦安汇 vs CC Pairlet

锦安汇是一款完全离线运行的图片视频处理工具箱，所有处理在设备本地完成，保护您的隐私安全。 图片工具 • 图片压缩：自动/自定义双模式，可调尺寸与质量 • 图片拼接：竖排/横排/宫格三种布局，可调间距、背景色、圆角 • 图片水印：支持文字水印与图片水印 • EXIF 查看与清除：查看拍摄参数等信息，一键清除隐私信息 • 格式转换：JPEG / PNG / HEIC 互转，批量处理 视频工具 • 视频压缩：三种预设 + 自定义码率，实时进度 • 视频截取：精确选取片段，逐帧微调，导出您需要的视频片段 • 视频转 GIF：视频片段转 GIF 动图，可调帧率与画质

Keep coding work moving when you are away from the keyboard. CC Pairlet (formerly CC Pocket) is an open-source, local-first companion for command-line AI coding agents. Your agent keeps running on your own Mac, Windows, or Linux computer, against your own checkout. From your iPhone or iPad, you can watch it work, answer permission requests, continue the same session, and review what changed. SUPPORTED AGENTS • Claude Code • Codex • OpenCode • Kimi Code (Preview) • ZCode • DeepSeek Harness (narrow support) WHAT YOU CAN DO • See projects, sessions, live output, tool calls, and background work • Allow or deny supported permission requests; unanswered requests fail closed • Send prompts, images, and voice input • Continue the same task without starting over • Review changed files and line-level diffs where the selected agent exposes them • Pair multiple computers and switch between them LOCAL-FIRST BY DESIGN The daemon runs on your computer and drives the agent CLI there. It makes an outbound connection, so you do not need to expose an inbound port. App-to-daemon traffic is end-to-end encrypted; the hosted relay only forwards ciphertext and does not hold your message content or private keys. No Pairlet account is required. You can also self-host the relay. HONEST CAPABILITY BOUNDARIES Agent backends are not equivalent. Approval, changed-files, usage, model controls, and other features depend on what each agent exposes. OpenCode has no enforceable interactive approval and runs with Full access. DeepSeek support is deliberately narrow: remote approvals fail closed, and changed-files, usage, and model switching are not available. Pairlet is MIT-licensed and built in Kotlin. Website and user guide: https://pairlet.org/en/ Source and security model: https://github.com/heypandax/pairlet