perplexity-web-mcp CLAUDE.md

Instructions for Perplexity Web MCP, a tool that lets programs and coding agents ask Perplexity AI questions through a terminal command or an MCP server. Perplexity AI can search the web and return answers with sources.

In plain words
What is it for?
Use them to install and authenticate the tool, ask web-based questions from a terminal, run the MCP server, or work on its client, models, rate limits, and stored login token.
Why use it?
They explain installation, login, configuration, and querying, removing the need to work out how the command-line and server interfaces connect.

Instructions file

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 instructions/devdotbo/perplexity-web-mcp/claude-md
Clone the repo
git clone --depth 1 https://github.com/devdotbo/perplexity-web-mcp
Per session 736 This file is loaded in full into every session.
When invoked 736 The same file — it is already loaded in full.
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 $0.00736 $0.00736
Opus 5 $0.00368 $0.00368
Sonnet 5 $0.00147 $0.00147
Haiku 4.5 $0.00074 $0.00074

Measured 2d ago against content hash 4297c37b191b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

perplexity-web-mcp CLAUDE.md 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 2d ago.

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.

CLAUDE.md · 87 lines

How it starts

The opening of the file, as written. The whole thing — 87 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Perplexity Web MCP

CLI, MCP server, and API-compatible interface for Perplexity AI's web interface.

Quick Start

# Install
uv venv && uv pip install -e .

# Authenticate
pwm login

# Query from terminal
pwm ask "What is quantum computing?"

# Run MCP server
pwm-mcp

Project Structure

src/perplexity_web_mcp/
├── __init__.py          # Package exports
├── shared.py            # Shared query logic (MODEL_MAP, ask(), used by CLI + MCP)
├── council.py           # Model Council (parallel multi-model queries + synthesis)
├── core.py              # Perplexity client, Conversation class
├── models.py            # Model definitions (GPT, Claude, Gemini, Grok, etc.)
├── config.py            # ClientConfig, ConversationConfig
├── enums.py             # CitationMode, SearchFocus, SourceFocus
├── http.py              # HTTP client with retry/rate limiting
├── rate_limits.py       # Rate limit checking via /rest/rate-limit/all
├── token_store.py       # Token persistence (~/.config/perplexity-web-mcp/token)
├── data/                # Bundled Agent Skill (SKILL.md + references/)
├── cli/
│   ├── main.py          # Unified CLI entry point (pwm)
│   ├── auth.py          # Authentication flow
│   ├── setup.py         # MCP server setup for AI tools
│   ├── skill.py         # Agent Skill management
│   ├── doctor.py        # Diagnostic checks
│   └── ai_doc.py        # --ai flag documentation
├── mcp/
│   └── server.py        # MCP server (imports from shared.py)
└── api/
    └── server.py        # Anthropic/OpenAI API compatibility

CLI Commands

pwm ask "query" [-m MODEL] [-t] [-s SOURCE]  # Query Perplexity
pwm council "query" [-m MODELS] [-t] [-s SOURCE]  # Model Council (multi-model)
pwm research "query" [-s SOURCE]              # Deep research
pwm login [--check] [--email E --code C]      # Authentication
pwm usage [--refresh]                          # Rate limits
pwm setup [list|add|remove] CLIENT             # MCP config
pwm skill [list|install|uninstall] TOOL        # Skill management
pwm doctor [-v]                                # Diagnostics
pwm --ai                                       # AI reference doc

Read the full file on GitHub · 87 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. 2d ago First seen · 87 lines · 736 tokens per session scan A 4297c37b191b

Subscribe to this mod's changes

perplexity-web-mcp CLAUDE.md is an instructions file published in the GitHub repository devdotbo/perplexity-web-mcp (15 stars, last pushed 4mo ago), licensed MIT. It adds 736 tokens to every session, about $0.0037 per session on Opus 5. 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.