# Pup — Datadog CLI Companion for AI Agents

> Pup is an Apache-2.0 Datadog CLI with 200+ commands across 33+ products, so agents can query logs, metrics, RUM, and security data via OAuth login.

## Install

Copy the content below into your project:

## Quick Use

```bash
brew install datadog-labs/pack/pup
pup auth login
pup logs search --query='status:error' --from='1h'
pup metrics query --query='avg:system.cpu.user{*}'
```

## Intro

Pup is an Apache-2.0 Datadog CLI with 200+ commands across 33+ products, so agents can query logs, metrics, RUM, and security data via OAuth login.

**Best for:** SRE/DevOps teams letting agents query Datadog safely with scoped auth and structured outputs

**Works with:** Datadog accounts + OAuth login, Homebrew installs, JSON/YAML output consumers

**Setup time:** 3-6 minutes

### Key facts (verified)

- GitHub: 773 stars · 72 forks · pushed 2026-05-12.
- License: Apache-2.0 · owner avatar + repo URL verified via GitHub API.
- README-verified entrypoint: `brew install datadog-labs/pack/pup`.

## Main

- Use Pup for agent-first observability: it exposes a discoverable CLI surface that’s easier for agents than browsing docs.

- Prefer scoped OAuth auth for safety, and keep agent credentials separated from human admin credentials.

- Wrap common queries as runbooks/scripts so your agents can follow repeatable investigation playbooks.

### Source-backed notes

- README describes 200+ commands across 33+ Datadog product domains.
- README highlights OAuth2 + PKCE for scoped access rather than long-lived keys.
- README Quick Start shows `pup auth login` and example queries for logs and metrics.

### FAQ

- **How do agents authenticate?**: README shows OAuth2 + PKCE via `pup auth login`.
- **Is output machine-friendly?**: Yes—README mentions structured JSON/YAML output.
- **How many APIs does it cover?**: README claims 200+ commands across 33+ product domains.

## Source & Thanks

> Source: https://github.com/DataDog/pup
> License: Apache-2.0
> GitHub stars: 773 · forks: 72

---

<!-- ZH -->

## Quick Use

```bash
brew install datadog-labs/pack/pup
pup auth login
pup logs search --query='status:error' --from='1h'
pup metrics query --query='avg:system.cpu.user{*}'
```

## Intro

Pup 是 Apache-2.0 许可的 Datadog CLI，覆盖 33+ 产品与 200+ 命令；通过 OAuth 登录让 agent 可按权限查询日志、指标、RUM 与安全数据，用于故障排查与可观测性运维自动化。

**Best for:** 希望让 agent 以权限受控方式安全调用 Datadog 的 SRE/DevOps 团队

**Works with:** Datadog 账号 + OAuth 登录、Homebrew 安装、可消费 JSON/YAML 输出的系统

**Setup time:** 3-6 minutes

### Key facts (verified)

- GitHub：773 stars · 72 forks；最近更新 2026-05-12。
- 许可证：Apache-2.0；作者头像与仓库链接均已通过 GitHub API 复核。
- README 中核对过的入口命令：`brew install datadog-labs/pack/pup`。

## Main

- 用 Pup 做 agent-first 的可观测性：CLI 的可发现命令面更适合 agent 自动探索与调用。

- 优先用 OAuth 的权限范围控制，并将 agent 凭证与人类管理员凭证隔离。

- 把常用查询封装成 runbook/脚本，让 agent 按可复现的排障剧本执行。

### Source-backed notes

- README 描述其覆盖 33+ Datadog 产品域与 200+ 命令。
- README 强调使用 OAuth2 + PKCE 做权限受控访问，而不是长期 token。
- README Quick Start 给出 `pup auth login`，并提供日志/指标查询示例。

### FAQ

- **agent 如何登录？**：按 README，用 `pup auth login` 走 OAuth2 + PKCE。
- **输出适合机器解析吗？**：是。README 提到结构化 JSON/YAML 输出。
- **覆盖范围有多大？**：README 声称覆盖 33+ 产品域与 200+ 命令。

## Source & Thanks

> Source: https://github.com/DataDog/pup
> License: Apache-2.0
> GitHub stars: 773 · forks: 72


---
Source: https://tokrepo.com/en/workflows/pup-datadog-cli-companion-for-ai-agents
Author: Script Depot