pi

Setup instructions for Webcmd, a command-line tool for browser work, so the Pi coding agent can discover and use its skills. Pi is a terminal-based coding assistant.

In plain words
What is it for?
Use it to install Webcmd, check its browser runtime, add Webcmd skills to Pi, and choose where those skills are stored.
Why use it?
It brings the required installation, configuration, project trust, and browser checks into one setup process. The supplied source document must be fetched before following the full instructions.

Agent

Install

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.

agentmods
npx agentmods add agents/agentrhq/webcmd/pi
Clone the repo
git clone --depth 1 https://github.com/agentrhq/webcmd
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,202 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00000 $0.01202
Opus 5 $0.00000 $0.00601
Sonnet 5 $0.00000 $0.00240
Haiku 4.5 $0.00000 $0.00120

Measured yesterday against content hash 4fa64e094012, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

pi 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 yesterday.

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.

docs/agents/pi.md · 76 lines

How it starts

The opening of the file, as written. The whole thing — 76 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Agent prompt

Fetch and follow https://raw.githubusercontent.com/agentrhq/webcmd/main/start.md to set up Webcmd end to end.

Manual

Requirements

  • Node.js 20.6+
  • The webcmd npm CLI, installed globally or in the project
  • Pi (@earendil-works/pi-coding-agent), running in a terminal session
  • A browser runtime; confirm with webcmd doctor before browser work

Install and configure

Install Webcmd and its skills:

npm install -g @agentrhq/webcmd
webcmd doctor
webcmd skills add

When webcmd skills add prompts, choose the agents provider. It links skills into ~/.agents/skills/ (user) or .agents/skills/ (project), both of which Pi scans for skills on startup alongside its own ~/.pi/agent/skills/ and .pi/skills/ directories. Pi then surfaces webcmd-usage and webcmd-browser as skills.

To install into Pi's own skill directories instead, pass a custom path:

webcmd skills add --path ~/.pi/agent/skills   # user level
webcmd skills add --path .pi/skills           # project level

Project skills under .pi/skills and .agents/skills load only after the project is trusted. Accept Pi's trust prompt on first run in the project, or set defaultProjectTrust to always in ~/.pi/agent/settings.json. Restart Pi after installing skills so the new skill descriptions are picked up at startup.

Override default tools

Pi ships seven built-in tools — read, write, edit, bash, grep, find, ls — and no web, search, or browser tool at all, so there is nothing built in to turn off. Webcmd works through bash as soon as the CLI is on PATH. Do not reach for --no-builtin-tools: it strips read/write/edit/bash, and bash is how Webcmd runs.

Pi is the harness where competing tools are most likely to come from what the user installed, since every web capability it has is an extension or skill. Check for these and ask before removing any:

Installed Overlaps What to do
pi-skills/browser-tools Browser automation over CDP (Chrome on :9222) Recommend removing, or tell Pi to prefer Webcmd
pi-agent-browser-native Exposes agent-browser as a native tool Recommend removing
pi-web-fetch Headless-Chrome fetch plus trafilatura extraction Recommend removing
pi-web-access Search and content extraction Keep the search half; set webSearch.enabled: false only if the user wants search gone too
pi-skills/brave-search Brave Search API, plus content extraction Keep it for search; tell Pi to prefer Webcmd for reading pages

Read the full file on GitHub · 76 lines

Changes

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.

  1. yesterday First seen · 76 lines · 0 tokens per session scan A 4fa64e094012

Subscribe to this mod's changes

pi is an agent published in the GitHub repository agentrhq/webcmd (1,594 stars, last pushed 2d ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 1,202 tokens. 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.