A tool that converts a web page or HTML you already have into Markdown for reading or supplying to an AI model. It preserves code blocks, tables, tab panels, image references, and page details such as the title and source.

Use it to read or save documentation, API references, release notes, articles, and README files as structured Markdown.

It helps when a normal web fetch returns only a summary, navigation, or a JavaScript shell instead of the page's useful text. It also avoids asking the agent to guess content it could not fetch.

page2ai — Skill for Claude Code by igorsaevets/page2ai-mcp

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.

Any agent

npx skills add igorsaevets/page2ai-mcp --skill page2ai

Clone the repo

git clone --depth 1 https://github.com/igorsaevets/page2ai-mcp

Made for: Claude Code.

Or install page2ai, the plugin that ships this one along with the rest of its 1 skill.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/igorsaevets/page2ai-mcp/page2ai/github.svg)](https://agentmods.dev/skills/igorsaevets/page2ai-mcp/page2ai)

Your own site

<a href="https://agentmods.dev/skills/igorsaevets/page2ai-mcp/page2ai"><img src="https://agentmods.dev/badge/skills/igorsaevets/page2ai-mcp/page2ai/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/igorsaevets/page2ai-mcp/page2ai"><img src="https://agentmods.dev/badge/skills/igorsaevets/page2ai-mcp/page2ai.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 100 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 1,688 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

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.

Model	Per session	Once invoked
Fable 5.1	$0.00100	$0.01688
Opus 5.5	$0.00040	$0.00675
Sonnet 5.5	$0.00020	$0.00338
Haiku 4.5	$0.00010	$0.00169

Measured 5d ago against content hash 6b8b2583cb93, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

page2ai 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 5d 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.

plugins/page2ai/skills/page2ai/SKILL.md · 70 lines

How it starts

The opening of the file, as written. The whole thing — 70 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Page2AI: web page to Markdown

The tool

page_to_markdown, served by the page2ai MCP server (this plugin starts it with npx, pinned to one page2ai-mcp release in the plugin's manifest).

claude.ai chat does not start local MCP servers, so there the tool is not among your tools. Say so instead of guessing a page's content, and use whatever fetch tool you do have.

Argument	Default	Meaning
`url`		Page to fetch (http or https; private, loopback and cloud-metadata addresses are refused)
`html`		Page HTML you already have, for example from a browser tool. Pass `url` with it so relative links resolve; nothing is fetched then
`include_frontmatter`	`true`	YAML block first: title, source, captured_at, language, canonical, extractor_version
`include_images`	`true`	Keep `![alt](src)` image references
`timeout_ms`	`15000`	1000 to 60000
`max_length`	`30000`	100 to 1000000. A longer page comes back in parts
`start_index`	`0`	Character offset to read from: the next part, or one section

Code blocks keep their language, tables stay tables, and tab panels come out as ### Tab: <label> sections instead of one merged block.

Picking the input

A public page: call it with url.
The result is a sign-in page, a cookie wall, an "enable JavaScript" shell or almost empty, or the call fails with HTTP 403 or 429: the server fetched without the user's cookies and ran no page scripts. The error's second line, or a note after the page, names the next step this server can offer; it knows whether Agent Bridge is set up.
- If a browser tool has that page open, take its HTML (document.documentElement.outerHTML) and call again with html and url.
- Otherwise ask the user once to set up Agent Bridge (below); after that, url pages open in their browser.
A refusal (403, 429, a CAPTCHA, a paywall) is the site's answer. Do not retry it with other headers, in a loop, or through another service.
Other errors say what to do as well: a 404 is a wrong address, a 429 means wait, a timeout means a larger timeout_ms, a PDF is not a page this tool reads.

Read the full file on GitHub · 70 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.

5d ago Changed · +5 lines 6b8b2583cb93
7d ago First seen · 65 lines · 100 tokens per session scan A 2020b4a5f8f1

Subscribe to this mod's changes

page2ai is a skill published in the GitHub repository igorsaevets/page2ai-mcp (0 stars, last pushed yesterday), licensed MIT. It adds 100 tokens to every session and 1,688 once invoked, about $0.0004 per session on Opus 5.5. 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 2026-10-31.

Documents LLM Research ai anthropic claude cursor html-to-markdown

page2ai next to 6 skills from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
page2aiigorsaevets/page2ai-mcp	Skill	0	yesterday	MIT	A	100	Claude Code
dcf-modelNousResearch/hermes-agent	Skill	262k	today	MIT	A	14	Claude Code, Codex
comps-analysisNousResearch/hermes-agent	Skill	262k	today	MIT	A	13	Claude Code, Codex
lbo-modelNousResearch/hermes-agent	Skill	262k	today	MIT	A	19	Claude Code, Codex
google-workspaceNousResearch/hermes-agent	Skill	262k	today	MIT	A	21	Claude Code, Codex
siyuanNousResearch/hermes-agent	Skill	262k	today	MIT	A	17	Claude Code, Codex
3-statement-modelNousResearch/hermes-agent	Skill	262k	today	MIT	A	18	Claude Code, Codex