netdiag-cli 0.1.0

One CLI for network troubleshooting: traceroute with BGP/ASN, DNS, latency, VPN checks, and SRE on-call presets.

netdiag

Production runbook CLI for network incidents - traceroute with ASN/BGP, DNS drift, latency, VPN checks, and on-call presets in one binary.

Stdlib-only at runtime · --json on every command · Exit codes built for automation

netdiag doctor
netdiag oncall <HOST> --json

---

Install

Method	Command
From source (dev)	python3 -m pip install -e ".[dev]"
PyPI	python3 -m pip install netdiag-cli
From Git	python3 -m pip install "git+https://github.com/akintunero/netdiag.git"

Use python3 -m pip (not bare pip) on macOS if pip is missing. The PyPI package is netdiag-cli; the command is still netdiag.

System tools: ping (required), plus traceroute, dig, and routing/socket tools as needed. Run netdiag doctor to see what is on your PATH.

---

Runbook workflow

Step	Command	Exit code
1. Pre-flight	netdiag doctor	0 ready · 2 missing required tools
2. Incident check	netdiag oncall <HOST> --json	0 pass · 1 fail · 2 error
3. Ticket artifact	netdiag report <HOST> -o incident.md	same

Exit code semantics are stable in v0.x - see docs/CLI_CONTRACT.md.

netdiag oncall 1.1.1.1 --json
netdiag oncall YOUR-SERVICE.example.com --json | jq '.steps[] | select(.ok==false)'

Tip: Run one command per line. Do not paste trailing # comments from docs into zsh/bash.

VPN and corporate DNS

netdiag vpn --corp intranet.company.com
netdiag oncall app.internal --corp app.internal --vpn --json

Optional config

Copy docs/config.example.toml to ~/.config/netdiag/config.toml:

[defaults]
corp_host = "app.internal.corp"

[oncall]
preset = "oncall"

Shell completions

eval "$(netdiag completion bash)"
eval "$(netdiag completion zsh)"

About the CLI

netdiag --info    # version, platform, repo links, wordmark banner

Set NETDIAG_NO_BANNER=1 to hide the banner.

---

Why operators use it

One CLI	Replaces ad-hoc dig + ping + curl + traceroute scripts
--json	PagerDuty, CI, Slack bots, jq filters
Presets	web, api, vpn, oncall - tuned check bundles
Small runtime	No required PyPI dependencies

---

Quick start (developers)

git clone https://github.com/akintunero/netdiag.git
cd netdiag
python3 -m pip install -e ".[dev]"
netdiag doctor
python3 -m pytest tests/ -q

Smoke test (live network, all subcommands):

bash scripts/cli_smoke.sh

Example probes:

netdiag oncall 1.1.1.1 --json
netdiag oncall 1.1.1.1 --preset api --json
netdiag trace 8.8.8.8 -m 12 --no-bgp-api
netdiag report 1.1.1.1 -o incident.md

---

Command reference

Command	Description
doctor	Pre-flight: required tools for runbooks
oncall	Primary incident bundle (preset oncall)
report	Markdown report (-o file.md, --vpn)
vpn	VPN, split-tunnel, DNS drift
check	Health check; --preset for bundles
trace	Traceroute + ASN/BGP enrichment
ping / latency	RTT, jitter, percentiles
dns / dns-trace / dns-compare / dns-all	DNS tooling
port / ports	TCP probes and scans
http / tls / redirects / headers	Application layer
speed	CDN throughput (Cloudflare by default)
compare	Two-host DNS + ping diff
presets	List available presets
completion	bash / zsh completions

Also: route, ifaces, local-ports, connections, whois, subnet, ip, ptr, mtr, and more - netdiag --help.

All commands support --json.

---

Documentation

Doc	Link
Docs index	docs/README.md
Cookbook (on-call recipes)	docs/cookbook.md
CLI contract (exit codes, flags)	docs/CLI_CONTRACT.md
Contributing	CONTRIBUTING.md
Security	SECURITY.md
Roadmap	ROADMAP.md
Changelog	CHANGELOG.md

---

Author

Olúmáyòwá Akinkuehinmi - @akintunero · [e-mail]

License

MIT - see LICENSE.

Responsible use

Use only on networks and hosts you are authorized to test.