pair-programmer

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

VideoDB Pair Programmer

AI pair programming with real-time screen and audio context. Record your screen and audio, with AI-powered indexing that logs visual and audio events in real-time.

具备实时屏幕和音频上下文的AI结对编程工具。可录制你的屏幕与音频，通过AI驱动的索引功能实时记录视觉和音频事件。

Commands

When user asks for a command, read the corresponding file for instructions:

Command	Description	Reference
`/pair-programmer record`	Start screen/audio recording	See commands/record.md
`/pair-programmer stop`	Stop the running recording	See commands/stop.md
`/pair-programmer search`	Search recording context (screen, mic, audio)	See commands/search.md
`/pair-programmer what-happened`	Summarize recent activity	See commands/what-happened.md
`/pair-programmer setup`	Install deps and configure API key	See commands/setup.md
`/pair-programmer config`	Change indexing and other settings	See commands/config.md

当用户询问命令时，请读取对应文件获取说明：

Command	Description	Reference
`/pair-programmer record`	开启屏幕/音频录制	参见 commands/record.md
`/pair-programmer stop`	停止正在进行的录制	参见 commands/stop.md
`/pair-programmer search`	搜索录制上下文（屏幕、麦克风、音频）	参见 commands/search.md
`/pair-programmer what-happened`	总结近期活动	参见 commands/what-happened.md
`/pair-programmer setup`	安装依赖并配置API key	参见 commands/setup.md
`/pair-programmer config`	修改索引及其他设置	参见 commands/config.md

How It Works

User runs
```
/pair-programmer setup
```
to install dependencies and set
```
VIDEO_DB_API_KEY
```
environment variable
User runs
```
/pair-programmer record
```
to start recording
A picker UI appears to select screen and audio sources
Recording starts and events are logged to
```
/tmp/videodb_pp_events.jsonl
```
User can stop recording from the tray icon (🔴 PP → Stop Recording)

用户运行
```
/pair-programmer setup
```
安装依赖并设置
```
VIDEO_DB_API_KEY
```
环境变量
用户运行
```
/pair-programmer record
```
开启录制
弹出选择器UI供用户选择屏幕和音频源
录制启动，事件会被记录到
```
/tmp/videodb_pp_events.jsonl
```
用户可以通过托盘图标停止录制（🔴 PP → 停止录制）

Output Files

Path	Content
`/tmp/videodb_pp_pid`	Process ID of the recorder
`/tmp/videodb_pp_events.jsonl`	All WebSocket events (JSONL format)
`/tmp/videodb_pp_info.json`	Current session info (session_id, rtstream_ids)

路径	内容
`/tmp/videodb_pp_pid`	录制器的进程ID
`/tmp/videodb_pp_events.jsonl`	所有WebSocket事件（JSONL格式）
`/tmp/videodb_pp_info.json`	当前会话信息（session_id、rtstream_ids）

Event File Format

Events are written as JSONL (one JSON object per line):

json

{"ts": "2026-03-05T10:15:30.123Z", "unix_ts": 1709374530.12, "channel": "visual_index", "data": {"text": "User is viewing VS Code with auth.ts open"}}
{"ts": "2026-03-05T10:15:31.456Z", "unix_ts": 1709374531.45, "channel": "transcript", "data": {"text": "Let me check the login flow", "is_final": true}}

事件以JSONL格式写入（每行一个JSON对象）：

json

{"ts": "2026-03-05T10:15:30.123Z", "unix_ts": 1709374530.12, "channel": "visual_index", "data": {"text": "User is viewing VS Code with auth.ts open"}}
{"ts": "2026-03-05T10:15:31.456Z", "unix_ts": 1709374531.45, "channel": "transcript", "data": {"text": "Let me check the login flow", "is_final": true}}

Environment Variables

The recorder reads these from environment variables:

Variable	Required	Description
`VIDEO_DB_API_KEY`	Yes	VideoDB API key
`VIDEO_DB_BASE_URL`	No	API endpoint (default: https://api.videodb.io)

录制器会从环境变量中读取以下配置：

变量名	必填	描述
`VIDEO_DB_API_KEY`	是	VideoDB API key
`VIDEO_DB_BASE_URL`	否	API端点（默认值：https://api.videodb.io）

Reading Context

Events are in

/tmp/videodb_pp_events.jsonl

. Use CLI tools to filter — never read the whole file.

Channel	Content	Density
`visual_index`	Screen descriptions	Dense (~1 every 2s)
`transcript`	Mic speech	Sparse (sentences)
`audio_index`	System audio summaries	Sparse (sentences)

Channel filter — use grep to filter by channel, pipe to

tail

for recent events:

bash

grep '"channel":"visual_index"' /tmp/videodb_pp_events.jsonl | tail -10

Keyword search — grep across all channels:

bash

grep -i 'keyword' /tmp/videodb_pp_events.jsonl

Time-window filter — filter events from the last N minutes:

Get current epoch:
```
$(date +%s)
```
Compute cutoff:
```
epoch - N*60
```
Filter lines where the
```
unix_ts
```
JSON field exceeds the cutoff
Pipe through
```
grep
```
to narrow by channel

Generate the appropriate filtering command (grep, awk, python3, jq) based on complexity.

For semantic search across indexed content, use

search-rtstream.js

bash

node search-rtstream.js --query="your query" --cwd=<PROJECT_ROOT>

<PROJECT_ROOT>
is the absolute path to the user's project directory. This is NOT the skill directory — resolve it before running the command.

See commands/search.md for the full search strategy and CLI patterns.

事件存储在

/tmp/videodb_pp_events.jsonl

中。请使用CLI工具进行过滤，请勿读取整个文件。

通道	内容	密度
`visual_index`	屏幕内容描述	高密度（约每2秒1条）
`transcript`	麦克风语音内容	低密度（按句子生成）
`audio_index`	系统音频摘要	低密度（按句子生成）

通道过滤 — 使用grep按通道过滤，通过管道传给

tail

获取最近的事件：

bash

grep '"channel":"visual_index"' /tmp/videodb_pp_events.jsonl | tail -10

关键词搜索 — 跨所有通道grep搜索：

bash

grep -i 'keyword' /tmp/videodb_pp_events.jsonl

时间窗口过滤 — 过滤最近N分钟内的事件：

获取当前时间戳：
```
$(date +%s)
```
计算 cutoff 阈值：
```
epoch - N*60
```
过滤
```
unix_ts
```
JSON字段大于cutoff的行
通过管道传给
```
grep
```
按通道缩小范围

根据复杂度生成合适的过滤命令（grep、awk、python3、jq）。

如需对索引内容进行语义搜索，请使用

search-rtstream.js

：

bash

node search-rtstream.js --query="your query" --cwd=<PROJECT_ROOT>

<PROJECT_ROOT>
是用户项目目录的绝对路径。它不是当前skill的目录，请在运行命令前解析该路径。

完整的搜索策略和CLI使用模式参见 commands/search.md。