Pp Archive Is

<!-- GENERATED FILE — DO NOT EDIT.
     This file is a verbatim mirror of library/media-and-entertainment/archive-is/SKILL.md,
     regenerated post-merge by tools/generate-skills/. Hand-edits here are
     silently overwritten on the next regen. Edit the library/ source instead.
     See the repository agent guide, section "Generated artifacts: registry.json, cli-skills/". -->

archive.today — Printing Press CLI

Prerequisites: Install the CLI

This skill drives the archive-is-pp-cli binary. You must verify the CLI is installed before invoking any command from this skill. If it is missing, install it first:

1. Install via the Printing Press installer. It defaults binaries to $HOME/.local/bin on macOS/Linux and %LOCALAPPDATA%\Programs\PrintingPress\bin on Windows:

   npx -y @mvanhorn/printing-press-library install archive-is --cli-only

2. Verify: archive-is-pp-cli --version
3. Ensure the reported install directory is on $PATH for the agent/runtime that will invoke this skill.

If the npx install fails (no Node, offline, etc.), fall back to a direct Go install (requires Go 1.26.6 or newer):

go install github.com/mvanhorn/printing-press-library/library/media-and-entertainment/archive-is/cmd/archive-is-pp-cli@latest

If --version reports "command not found" after install, the runtime cannot see the binary directory on $PATH. Do not proceed with skill commands until verification succeeds.

When to Use This CLI

Reach for this whenever a user wants to archive a URL, read a paywalled article, check whether something was previously archived, or batch-capture a list of URLs for research. Specifically good when:

• A user sends a paywalled link and asks "can you read this" → read fetches text via archive
• They want to preserve a URL that might change → save forces a fresh capture
• They want historical versions → history lists all known snapshots
• They have 20+ URLs to archive → bulk runs rate-limited batch archival

Don't reach for this if the URL is trivially scrapeable without archive services (no paywall, robots-allowed, direct HTTP works), or if the user wants the original source rather than a cached version.

Unique Capabilities

The whole CLI is unique — archive.today has no official API. But within this CLI, certain commands are the differentiators.

The hero commands

• read <url> — Find or create an archive for a URL. Looks up existing snapshots first (Memento timegate → CDX fallback); submits a fresh capture only if nothing exists. The "always do the right thing" command.

  This is how 90% of agent calls should start. It's idempotent — calling it twice on the same URL doesn't double-submit.

• get <url> [--raw] / tldr <url> — Fetch article text, optionally LLM-summarized. Automatic Wayback fallback when archive.today serves a CAPTCHA (which happens daily to cloud IPs).

  tldr pipes the fetched text through a summarization step — useful for agent chains where you want a short take without shipping 20KB of HTML back.

Durability operations

• save <url> — Force a fresh capture via /submit/?url=<x>&anyway=1. Use when read returns an existing snapshot that's too old or missing a paywall update.

• history <url> — List all known snapshots via Memento timemap parsing. Shows every capture date across both archive.today and Wayback.

• bulk [file] — Rate-limited batch archiving from a file or stdin. Reads URLs one per line, submits each with backoff, returns a report of successes / failures / pre-existing.

  grep -oE 'https?://[^ )]+' notes.md | archive-is-pp-cli bulk - archives every URL in a markdown file.

• request <url> — Fire-and-forget submit with optional wait+poll. Useful for long captures where you want to come back later.

Observability

• snapshots newest <url> — Just the newest snapshot URL for a target, useful in scripts.

• captures — List your local capture index (post-sync).

• feeds — archive.today's global recent-archives feed.

• --backend archive-is,wayback — Every read/get accepts a backend preference. Defaults to archive-is with Wayback fallback; flip the order for Wayback-primary.

Command Reference

Archive + retrieve:

• archive-is-pp-cli read <url> — Find or create (hero command)
• archive-is-pp-cli get <url> — Fetch article text (with Wayback fallback)
• archive-is-pp-cli tldr <url> — Fetch + summarize
• archive-is-pp-cli save <url> — Force fresh capture
• archive-is-pp-cli request <url> — Fire-and-forget submit
• archive-is-pp-cli request check <url> — Check whether a submitted request has completed

Listing + history:

• archive-is-pp-cli history <url> — All known snapshots
• archive-is-pp-cli snapshots <url> — Best known snapshot for a URL
• archive-is-pp-cli snapshots newest <url> — Newest snapshot URL
• archive-is-pp-cli snapshots timemap <url> — Raw Memento timemap
• archive-is-pp-cli captures — Local capture index
• archive-is-pp-cli feeds — Global recent feed

Batch:

• archive-is-pp-cli bulk [file] — Batch from file or stdin

Local store:

• archive-is-pp-cli sync / export / import / workflow archive — Local SQLite ops

Auth + health:

• archive-is-pp-cli auth — Config (no API key needed; auth is a no-op)
• archive-is-pp-cli doctor — Verify backend reachability

Recipes

Read a paywalled article

archive-is-pp-cli read "https://www.wsj.com/articles/..." --agent
# or: return just the text
archive-is-pp-cli get "https://www.wsj.com/articles/..." --agent

read returns the archive URL (finding existing or creating new). get returns extracted article text by default, falling back to Wayback if archive.today CAPTCHAs; add --raw only when you need the archived HTML.