Claude Review: What It Does, Pricing, and Alternatives

2026-05-22 KST first draft; promoted to content_status = qa_passed after the 2026-05-22 claude.com/pricing page-body read. Last source-freshness refresh 2026-06-24 KST. Generated from templates/tool-page-template.md. Passed Section A of qa/adsense-seo-quality-gate.md. Meta description (≤ 155 chars): Claude is Anthropic's conversational AI built for careful reasoning, long-document work, and developer workflows — here is what it does, who it fits, and where it falls short.

Quick verdict

• Best for: long-document analysis, reasoning-heavy writing, and coding assistance for people who want a careful, instructable assistant rather than a maximally chatty one.
• Not ideal for: users who want a single tool that also handles image generation, video, or deep multimodal product integrations out of the box.
• Pricing model: freemium with paid consumer tiers and a separate developer API. Plan names and monthly prices verified against the official pricing page on 2026-05-22.
• Free plan: yes — Anthropic lists a Free plan at $0 on claude.com/pricing as of 2026-05-22; specific message quotas and per-model availability still shift between releases and should be reconfirmed before quoting numbers older than ~90 days.
• Last verified: 2026-05-22 (claude.com/pricing page-body read)

Latest update (2026-05-31)

Anthropic has announced Claude Opus 4.8, described as the newest version of its most capable model. Per Anthropic's official announcement ("Introducing Claude Opus 4.8"):

• Available now at the same price. Opus 4.8 launched the same day it was announced and, per Anthropic, is offered at the same price as the previous version. (Plan inclusions and per-tier model availability still change over time — confirm on the official pricing page before relying on specifics. The Free/Pro/Max prices quoted lower on this page are from the 2026-05-22 read and were not re-verified for this update.)
• Effort control on claude.ai. Users on claude.ai can adjust how much effort the model puts into a response.
• Dynamic workflows in Claude Code. Claude Code adds dynamic workflows aimed at very large-scale problems.
• A faster, cheaper fast mode. Anthropic describes a fast mode for Opus 4.8 that runs at roughly 2.5× speed and is about three times cheaper than previous models.
• Honesty caveat. Anthropic says Opus 4.8 is built to be more honest — flagging uncertainty and being less likely to make unsupported claims — and cites its own evaluation that the model is around four times less likely than its predecessor to let flaws in its written code go unremarked. This is Anthropic's internal evaluation, not an independent benchmark, so treat it as a vendor claim rather than a settled ranking. It also does not eliminate the hallucination and verification caveats below.

This is a source-backed freshness note drawn from Anthropic's own announcement. We do not independently benchmark models or claim one tool beats another; verify current model availability and pricing on official sources before relying on them.

Source-freshness note (2026-06-24)

If your buying question is specifically Claude Code as an agent/workflow surface — a coding agent you run where you already work, rather than a separate chat window — a 2026-06-23 KST official-source gate confirms Anthropic's own pages for evaluating that yourself are reachable. The Claude Code product page (anthropic.com/claude-code, which resolves to claude.com/product/claude-code, titled "Claude Code by Anthropic | AI Coding Agent, Terminal, IDE", H1 "Claude Code", HTTP 200) and the official Claude Code documentation (docs.anthropic.com/en/docs/claude-code/overview, which resolves to code.claude.com/docs/en/overview, titled "Overview - Claude Code Docs", H1 "Overview", HTTP 200) both loaded in the same pass.

• Read the docs, not just the landing page. The product page tells you how the vendor frames Claude Code; the official documentation overview is where you confirm how it actually installs, runs, and fits into a terminal/IDE workflow. Evaluate both against how your team edits and ships code.
• Agent output is a draft, not a merge. A coding agent proposes edits and runs commands; a human still reviews, tests, and approves before anything merges or ships. The vendor's positioning is not a guarantee about any specific run.
• Vendor evidence only. These are Anthropic's own product and documentation pages, confirming the Claude Code agent/workflow surface is reachable — not an independent ranking or benchmark. Confirm current pricing, plan availability, model availability, and security/enterprise terms on the official pages before relying on specifics.

Source-backed freshness note: a reachability/title/H1 recheck of Anthropic's own Claude Code product and documentation pages from the 2026-06-23 KST agent-workflow official-source gate. No price, quota, per-plan feature, model-availability, benchmark, ranking, speed, accuracy, or superiority claim is made here; volatile specifics are routed to the official site, and vendor positioning is reported as vendor evidence only.

Source-freshness note (2026-06-18)

If you are weighing Claude as a combined chat assistant + coding-agent workflow, a 2026-06-18 KST recheck confirms Anthropic's own surfaces for evaluating that decision yourself are reachable and stable. The Claude Code product page (claude.com/product/claude-code, titled "Claude Code by Anthropic | AI Coding Agent, Terminal, IDE", HTTP 200), the pricing page (claude.com/pricing, titled "Plans & Pricing | Claude by Anthropic", HTTP 200), and the Anthropic newsroom (anthropic.com/news, titled "Newsroom \ Anthropic", HTTP 200) all loaded as full pages in the same pass.

• Evaluate both jobs at the source. Claude spans an everyday chat assistant and a terminal/IDE coding agent. Read the product, pricing, security, and enterprise surfaces on Anthropic's official pages and decide whether each fits how your team actually works — don't infer current details from this page.
• Repository / code-agent output still needs human review. A coding agent proposes edits and runs commands; treat its output as a draft your team reviews, tests, and approves before it merges or ships. The vendor's positioning is not a guarantee about any specific run.
• Vendor evidence only. These are Anthropic's own pages, confirming the product/pricing/newsroom surfaces are reachable — not an independent ranking or benchmark. Confirm current pricing, plan availability, model availability, and security/enterprise terms on the official pages before relying on specifics.

Source-backed freshness note: a reachability/title recheck of Anthropic's own Claude Code, pricing, and newsroom pages on 2026-06-18 KST. No price, quota, per-plan feature, model-availability, benchmark, ranking, speed, or superiority claim is made here; volatile specifics are routed to the official site, and vendor positioning is reported as vendor evidence only.

Source-freshness note (2026-06-12)

If you are weighing Claude as a developer coding-agent workflow rather than a chat assistant, note that Anthropic maintains a first-party Claude Code product surface — a coding agent positioned to work in the terminal and IDE — on its official product page (anthropic.com/claude-code, titled "Claude Code by Anthropic | AI Coding Agent, Terminal, IDE", read HTTP 200 on 2026-06-12 KST).

• Workflow fit. Claude Code is a terminal/IDE coding-agent surface — it runs where developers already work, rather than only as a separate chat window. Evaluate it against how your team actually edits and ships code.
• Repository changes need human review. A coding agent proposes edits to your repository; those changes still need a human to review, test, and approve before they merge or ship. Treat agent output as a draft, not a final commit.
• Vendor evidence only. The page shows Anthropic offers a first-party coding-agent surface with developer- and enterprise-oriented positioning. That is the vendor's own framing, not an independent ranking — confirm current pricing, plan availability, and model availability on Anthropic's official pricing and product pages before relying on specifics.

Source-backed freshness note drawn from Anthropic's own product page. No benchmark, ranking, price, quota, speed, model-availability, or superiority claim is made here; vendor positioning is reported as vendor evidence only.

Source-freshness note (2026-06-15)

A 2026-06-15 KST recheck confirms Anthropic's official surfaces for verifying Claude yourself are reachable and stable. The pricing page (claude.com/pricing, titled "Plans & Pricing | Claude by Anthropic", HTTP 200; the older anthropic.com/pricing link resolves to the same page) and the product overview (claude.com/product/overview, HTTP 200) both load as full pages, and the Claude Code product link now resolves to claude.com/product/claude-code.

• Verify pricing and plans at the source. Plan names, monthly prices, message limits, and which models each tier includes change between releases. Read them on Anthropic's official pricing page rather than trusting any figure — including the ones quoted lower on this page — once it is more than ~90 days old.
• Workflow fit over headline specs. Choose Claude by how it slots into your actual work — long-document reading, instructable drafting, or terminal/IDE coding — not by a single benchmark or version number. The right tier is the one whose surfaces and limits match your day.
• Keep a human in the loop. Treat any Claude output — prose, analysis, or code an agent proposes — as a draft to review, test, and verify before you rely on it or ship it. The vendor's safety positioning is not a guarantee about any specific response.

Source-backed freshness note: a reachability/title recheck of Anthropic's own pricing and product pages on 2026-06-15 KST. No new price, quota, benchmark, ranking, speed, model-availability, or superiority claim is made here; volatile specifics are routed to the official site.

What is Claude?

Claude is the conversational AI product line from Anthropic, an AI safety company. The product is offered through a web app, mobile apps, desktop clients, and a developer API. Anthropic positions Claude around careful reasoning, long-context comprehension, and instructable behavior, and the official product page lives on the claude.com domain at /product/overview (reachable via redirect from anthropic.com/claude).

The Claude product line includes multiple model variants released over time (Anthropic publicly references "Opus", "Sonnet", and "Haiku" naming for different tradeoffs of capability, speed, and cost). The exact lineup, version numbers, context-window sizes, and per-model availability shift between releases and should always be checked against Anthropic's own product and pricing pages before being quoted.

• Vendor: Anthropic
• Official homepage: https://claude.com/product/overview
• Category: AI Assistants (secondary: AI Writing & Editing, AI Coding Assistants)

Main use cases

• Use case 1 — Long-document analysis: uploading or pasting long PDFs, contracts, research papers, or codebases and asking Claude to summarize, extract, cite, or compare across sections. Claude's positioning around long-context work makes this the canonical "what is Claude actually for" answer for many knowledge workers.
• Use case 2 — Reasoning-heavy writing and editing: drafting analytical memos, structured arguments, code reviews, or technical explanations where you want the model to follow instructions tightly and explain its reasoning. Claude's instructable behavior is often described by users as easier to steer toward "do not invent facts" than a more eager generalist assistant.
• Use case 3 — Coding assistance and technical problem solving: explaining unfamiliar code, generating boilerplate, walking through refactors, or running through a debugging dialogue. Anthropic explicitly markets Claude for developer workflows alongside a public developer API.

Pricing and plans

Anthropic offers consumer plans through the Claude web and mobile apps and a separate metered developer API. The values below were read directly from claude.com/pricing on 2026-05-22 KST. Plan structures shift; reconfirm with the official page before quoting numbers older than ~90 days.

• Free — $0 for everyone (entry tier; quota and per-model access can change between releases).
• Pro — $20/month billed monthly, or $17/month with annual billing.
• Max 5x — "From $100/month" (higher usage allowance vs Pro).
• Max 20x — also starts From $100/month at a higher tier of usage allowance within the Max offering.
• Developer API — listed as a separate pricing surface from the consumer plans; specific per-model token rates were not in scope of this fetch and should be read directly from the API pricing page.

Source: live page-body read of https://claude.com/pricing on 2026-05-22 KST. Specific message limits, per-model availability, team/enterprise seat pricing, and API per-token rates were not asserted in this fetch and should be re-read directly when needed.

Pros

• Strong fit for tasks involving long documents, multi-step reasoning, or structured analysis, given Anthropic's public positioning and product framing.
• Available through web, iOS, Android, desktop, and API surfaces, which makes it easy to bring into multiple workflows without locking into a single platform.
• Anthropic publishes a "constitutional AI" and safety stance and provides documentation about its model behavior policies — useful for users who want to read the vendor's posture before relying on it.

Cons and caveats

• Anthropic's safety positioning is the vendor's public stance, not a guarantee about any specific output. Claude can still hallucinate, miss instructions, or refuse benign tasks.
• Plan structure and pricing have changed multiple times across the Claude product line. Numbers older than ~90 days should not be trusted without rechecking.
• Image generation, video generation, and deep third-party app integrations are not the core focus of the consumer Claude product line, even when adjacent features ship.
• Data-handling differs between consumer plans, team plans, and the API. Treat Anthropic's published data policy as the only authoritative source on what is or is not used for model training.
• Outputs are not professional advice. Do not treat Claude responses as a substitute for licensed medical, legal, accounting, or financial counsel.

Alternatives

• ChatGPT — better if you want the largest mainstream ecosystem of plugins, custom GPTs, third-party tools, and tutorial content, or if your team is already standardized on OpenAI's APIs.
• Gemini — better if you are deep inside Google Workspace (Docs, Gmail, Drive, Calendar) and you would like the assistant to work natively alongside those apps.
• Microsoft Copilot — better if your organization is standardized on Microsoft 365 and Windows, and you want the assistant to appear inside Word, Excel, Outlook, and Teams as a native surface.

Buyer control and the review boundary

If you are evaluating Claude as a buyer rather than a casual chat user, the deciding question is less "can it reason and write?" than "which of your workspace, code, and context can Claude read or write, and where does the review boundary sit before its output becomes a record or a commit?" These are the durable control questions to ask. <span id="claude-review-boundary-2026-06-27"></span>

• Control what Claude can read and write. Claude spans a chat window, uploaded long documents, and — through Claude Code — a terminal/IDE coding agent that proposes edits and runs commands. Decide up front which files, repositories, and context each surface may read from and write into before you adopt it across the team, rather than granting broad access by default.
• Human review before output becomes a record. Treat any Claude response — prose, analysis, or an agent's proposed edits — as a draft, not a record. Name who reviews, tests, and approves it before it becomes a published document, a system-of-record decision, or merged code. The vendor's safety positioning is not a guarantee about any specific output, as Cons and caveats describes.
• Keep the approval boundary on the human side. A coding agent runs commands and a chat assistant drafts confidently, but neither owns whether the result is correct. Keep approval a hard gate before a draft, summary, or change ships — the closer the work is to production code or a shared record, the tighter that gate should be.
• Route the comparison to the right page. When the real question is Claude versus another tool for a specific job, follow the workflow-fit decision paths below — for example ChatGPT vs Claude or Claude vs Replit AI — or browse the AI Assistants category to scope the field before deciding who controls what.

Evergreen decision framing only. No price, plan, quota, model-availability, benchmark, ranking, speed, accuracy, superiority, or legal claim is made here; verify current plan inclusions, data-handling, and security/enterprise terms on Anthropic's official pages, and confirm how each Claude surface fits your team's review process against your own practice.

Where to compare Claude next

If Claude is already on your shortlist, the next question is usually "against what, and for which job?" These side-by-side pages are organized by workflow fit, not by a winner — each one walks through where one tool's shape suits a particular task better than the other, so you can follow the path that matches your own work:

• ChatGPT vs Claude — when you are weighing a broad mainstream generalist against an assistant tuned for long-document analysis and tightly-followed instructions.
• Claude vs Gemini — when the deciding factor is how much of your day already lives inside Google Workspace versus a standalone reasoning assistant.
• Claude vs Microsoft Copilot — when your organization is standardized on Microsoft 365 and the question is in-suite integration versus a standalone chat surface.
• Claude vs Replit AI — when your task is shipping code and you are choosing between an in-editor/IDE coding assistant and a general reasoning-and-coding chat.
• Zapier AI vs Claude — when your job leans toward connecting apps and automating multi-step workflows rather than holding a conversation with one assistant.

To browse the whole field rather than a single head-to-head, start from the AI Assistants category. These links are decision paths, not rankings — no benchmark, price, quota, speed, or model-availability claim is made here; the comparison pages route any such specifics to the official sources.

Who should not use Claude

• Teams that need a single tool that also does production-quality image, video, or audio generation today.
• Users whose top requirement is real-time web research with inline citations from many sources — a dedicated AI answer engine is closer to that job.

Author selection rubric

Choose Claude when at least two of these are true:

• Your daily work involves reading or writing documents longer than ten pages.
• You care that the assistant reliably follows your instructions over many turns.
• You want a developer-friendly API alongside an end-user chat app.

Avoid Claude when any of these are true:

• Your team already pays for a comparable assistant that is more tightly integrated with your existing workflow tools and Claude would only duplicate it.
• Your single most important workflow is real-time multimodal generation across many media types.
• You require a vendor with a deeper formal procurement profile in your jurisdiction than Anthropic currently offers your organization.

Sources

• Official homepage: https://claude.com/product/overview — recorded as src-anthropic-claude-overview-2026-05-21 in data/sources.json; reached via redirect from anthropic.com/claude during prior scan.
• Official pricing page: https://claude.com/pricing — recorded as src-anthropic-claude-pricing-2026-05-22 in data/sources.json with access_status = ok after a 2026-05-22 page-body read; this is the source of every plan/price quoted on this page.
• Official announcement: https://www.anthropic.com/news/claude-opus-4-8 ("Introducing Claude Opus 4.8") — recorded as src-anthropic-claude-opus-4-8-2026-05-31 in data/sources.json with access_status = ok after a 2026-05-31 read; source of the "Latest update" section (Opus 4.8 release, effort control, Claude Code dynamic workflows, fast-mode framing, honesty/uncertainty caveat) and nothing else.
• Official product page: https://www.anthropic.com/claude-code ("Claude Code by Anthropic | AI Coding Agent, Terminal, IDE") — recorded as src-anthropic-claude-code-2026-06-12 in data/sources.json with access_status = ok after a 2026-06-12 KST HTTP 200 read; source of the "Source-freshness note (2026-06-12)" section, supporting only that Anthropic offers a first-party Claude Code terminal/IDE coding-agent surface with developer/enterprise positioning. Used as vendor evidence only; no pricing, plan, model-availability, benchmark, ranking, speed, or superiority claim is drawn from it.
• Official pricing/product reachability recheck: https://claude.com/pricing ("Plans & Pricing | Claude by Anthropic") — recorded as src-anthropic-claude-pricing-2026-06-15 in data/sources.json with access_status = ok after a 2026-06-15 KST HTTP 200 read; source of the "Source-freshness note (2026-06-15)" section. Confirms reachability/final-URL/title and the Free/Pro/Max plan-tier names, plus companion HTTP 200 reachability of claude.com/product/overview and the anthropic.com/claude-code → claude.com/product/claude-code redirect, in the same pass. Reachability/title evidence only — no USD price, quota, model-availability, benchmark, ranking, speed, or superiority claim is drawn from it; the 2026-05-22 page-body read remains the source of every plan/price quoted on this page.
• Official Claude Code / pricing / newsroom reachability recheck (2026-06-18): https://claude.com/product/claude-code ("Claude Code by Anthropic | AI Coding Agent, Terminal, IDE"), https://claude.com/pricing ("Plans & Pricing | Claude by Anthropic"), and https://www.anthropic.com/news ("Newsroom \ Anthropic") — recorded as src-anthropic-claude-code-pricing-news-2026-06-18 in data/sources.json with access_status = ok after 2026-06-18 KST HTTP 200 reads; source of the "Source-freshness note (2026-06-18)" section. Reachability/title evidence only (vendor pages) — no USD price, quota, per-plan feature, model-availability, benchmark, ranking, speed, or superiority claim is drawn from it; the 2026-05-22 page-body read remains the source of every plan/price quoted on this page. This is not a new revenue page and the revenue inventory is unchanged.
• Official Claude Code product + docs reachability recheck (2026-06-23 agent-workflow source gate): https://www.anthropic.com/claude-code (resolves to https://claude.com/product/claude-code, "Claude Code by Anthropic | AI Coding Agent, Terminal, IDE", H1 "Claude Code") and https://docs.anthropic.com/en/docs/claude-code/overview (resolves to https://code.claude.com/docs/en/overview, "Overview - Claude Code Docs", H1 "Overview") — recorded as src-anthropic-claude-code-workflow-source-gate-2026-06-23 in data/sources.json with access_status = ok after 2026-06-23 KST HTTP 200 reads logged in data/agent-workflow-source-gate-2026-06-23-1007.json; source of the "Source-freshness note (2026-06-24)" section, supporting only that Anthropic's first-party Claude Code agent/workflow product and documentation surfaces are reachable. Reachability/title/H1 evidence only (vendor pages) — no USD price, quota, per-plan feature, model-availability, benchmark, ranking, speed, accuracy, or superiority claim is drawn from it; the 2026-05-22 page-body read remains the source of every plan/price quoted on this page. This is not a new revenue page and the revenue inventory is unchanged.
• Vendor: Anthropic — https://www.anthropic.com/

Internal links (at least 3)

• Category page: /ai-assistant/
• Alternative tool: /tools/chatgpt/
• Comparison pages: /compare/chatgpt-vs-claude/, /compare/claude-vs-gemini/, /compare/claude-vs-microsoft-copilot/, /compare/claude-vs-replit-ai/, /compare/zapier-ai-vs-claude/

Disclosure

• Affiliate links: none.
• Sponsored content: none. Anthropic has no relationship to this page.
• Generative AI assistance: this draft was assembled with the help of an AI assistant working from the HMP source records; all factual claims are constrained to what those records support.

Trademark notice

Claude and Anthropic are trademarks of Anthropic. Use here is referential only and does not imply endorsement, partnership, or affiliation.

Update log

• 2026-06-27 (buyer-control / review-boundary qualified-traffic increment — LOCAL source/content edit only, NOT a new revenue page/product and NOT a deploy): added one compact, evergreen "## Buyer control and the review boundary" section placed after "## Alternatives" and before "## Where to compare Claude next", carrying durable marker claude-review-boundary-2026-06-27 as a single <span id="claude-review-boundary-2026-06-27"></span> in the section's lead. It frames Claude buying around buyer control over which workspace/code/context each Claude surface (chat, uploaded documents, and the Claude Code terminal/IDE coding agent) can read or write, human review before any output becomes a published record or merged code, keeping the approval boundary on the human side, and routing tool-vs-tool questions to the existing comparison and /ai-assistant/ category pages already on this page. Existing page/source context only — no web fetch and no data/* edit; no new price/quota/plan/model-availability/ranking/benchmark/speed/accuracy/superiority/legal claim; no Gumroad/UTM/affiliate/sponsored/coupon/checkout link; data/tools.json and last_verified_at unchanged; content_type/content_status stay tool_page/qa_passed. Local-only increment; no build/deploy performed — live verify/build is handled separately by Hermes.
• 2026-05-22 (draft): first local draft created from templates/tool-page-template.md. content_status = drafted.
• 2026-05-22 (qa pass): live page-body read of https://claude.com/pricing added concrete Free/Pro/Max plan names and prices; data/sources.json updated (src-anthropic-claude-pricing-2026-05-22, access_status = ok); data/tools.json pricing_summary, confidence_score, and last_verified_at refreshed. Section A1/A2 of qa/adsense-seo-quality-gate.md now satisfied. content_status advanced to qa_passed.
• 2026-05-31 (freshness refresh): added the "Latest update (2026-05-31)" section from Anthropic's official Claude Opus 4.8 announcement (src-anthropic-claude-opus-4-8-2026-05-31, access_status = ok) — Opus 4.8 release at the same price, effort control on claude.ai, Claude Code dynamic workflows, fast-mode framing, and the honesty/unsupported-claim caveat (Anthropic's own evaluation, not an independent benchmark). No pricing rows changed; the 2026-05-22 pricing read was not re-verified. This is a source-backed content refresh, not a new revenue page; no data/tools.json change.
• 2026-06-12 (source-freshness refresh — local source, NOT a new revenue page): added a compact "Source-freshness note (2026-06-12)" section after the "Latest update (2026-05-31)" section, framing Claude as a developer coding-agent workflow from Anthropic's official Claude Code product page (src-anthropic-claude-code-2026-06-12, https://www.anthropic.com/claude-code, title "Claude Code by Anthropic | AI Coding Agent, Terminal, IDE", access_status = ok after a 2026-06-12 KST HTTP 200 read). Covers workflow fit (terminal/IDE coding-agent surface), that repository changes still need human review before they ship, and that pricing/plan/model availability must be confirmed on official pages — vendor evidence only, no independent ranking. Added one source to data/sources.json and the claude sources_used. No pricing row changed; no benchmark, ranking, price, quota, speed, model-availability, or superiority claim added; no Gumroad/UTM/affiliate/sponsored/coupon link added; data/tools.json and last_verified_at unchanged.
• 2026-06-03 (internal-link/traffic refresh — live deployed, NOT a new revenue page): added a "Where to compare Claude next" section (placed after "## Alternatives", before "## Who should not use Claude") linking five now-live qa_passed Claude comparison pages — /compare/chatgpt-vs-claude/, /compare/claude-vs-gemini/, /compare/claude-vs-microsoft-copilot/, /compare/claude-vs-replit-ai/, /compare/zapier-ai-vs-claude/ — plus the /ai-assistant/ category, framed as workflow-fit decision paths rather than rankings. Updated the "Internal links" comparison row to list these five Claude comparisons (it previously listed only /compare/chatgpt-vs-claude/). No source was fetched and no volatile claim was added: no benchmark, ranking, price, quota, message limit, speed, or model-availability fact is asserted, and no commercial, affiliate, sponsored, or tracking-parameter CTA was added to this page. data/sources.json, data/tools.json, and last_verified_at are unchanged. Revenue inventory unchanged (18 tool + 51 comparison = 69 qa_passed pages). Hermes deployed the production build to Cloudflare Workers static assets; live /tools/claude/ returns HTTP 200 with the new marker and Gumroad/UTM 0.
• 2026-06-18 (source-freshness refresh — local source recheck, NOT a new revenue page): added a compact "Source-freshness note (2026-06-18)" section above the "Source-freshness note (2026-06-12)" section, framing Claude as a combined chat assistant + coding-agent workflow whose product/pricing/security/enterprise surfaces should be verified at Anthropic's official pages, with code-agent output still needing human review/testing and no independent ranking or benchmark. Evidence: a 2026-06-18 KST Hermes recheck found Anthropic's official surfaces reachable (HTTP 200) — https://claude.com/product/claude-code (title "Claude Code by Anthropic | AI Coding Agent, Terminal, IDE", 200000-byte body-sample SHA-256 c6b416f7aded1819…), https://claude.com/pricing (title "Plans & Pricing | Claude by Anthropic", SHA-256 93fc2b708cc9c68d…), and https://www.anthropic.com/news (final same URL, title "Newsroom \ Anthropic", SHA-256 4cec2ae7e8539e4c…) — recorded as src-anthropic-claude-code-pricing-news-2026-06-18 (access_status = ok) and added to the claude sources_used. Reachability/title evidence only: no exact price, quota, per-plan feature, benchmark, ranking, speed, or model-availability claim was added, and the 2026-05-22 page-body read remains the source of every plan/price quoted here. This is not a new revenue page and revenue inventory is unchanged; no Gumroad/UTM/affiliate/sponsored/coupon link added; data/tools.json and last_verified_at unchanged; still qa_passed, still 1 of the 18 tool pages.
• 2026-06-24 (source-freshness refresh — local source recheck, NOT a new revenue page): added a compact "Source-freshness note (2026-06-24)" section above the "Source-freshness note (2026-06-18)" section, framing Claude specifically as a Claude Code agent/workflow buying surface — read the official docs as well as the product landing page, treat agent output as a draft a human reviews/tests/approves before it merges, and confirm pricing/plan/model availability and security/enterprise terms on official pages. Evidence: the 2026-06-23 KST agent-workflow official-source gate (data/agent-workflow-source-gate-2026-06-23-1007.json) found two official Anthropic Claude Code surfaces reachable (HTTP 200) — https://www.anthropic.com/claude-code (final URL https://claude.com/product/claude-code, title "Claude Code by Anthropic | AI Coding Agent, Terminal, IDE", H1 "Claude Code", body-sample SHA-256 353d70aa436628b9…) and https://docs.anthropic.com/en/docs/claude-code/overview (final URL https://code.claude.com/docs/en/overview, title "Overview - Claude Code Docs", H1 "Overview", body-sample SHA-256 6a5576ea3caea682…) — recorded as src-anthropic-claude-code-workflow-source-gate-2026-06-23 (access_status = ok) and added to the claude sources_used. Reachability/title/H1 evidence only: no exact price, quota, per-plan feature, model-availability, benchmark, ranking, speed, accuracy, or superiority claim was added, and the 2026-05-22 page-body read remains the source of every plan/price quoted here. This is not a new revenue page and revenue inventory is unchanged; no Gumroad/UTM/affiliate/sponsored/coupon link added; data/tools.json and last_verified_at unchanged; still qa_passed, still 1 of the 18 tool pages. Local-only increment; live verify/build is handled separately.
• 2026-06-15 (source-freshness refresh — local source recheck, NOT a new revenue page): corrected the stale header (it still read Draft v0.1 / content_status = drafted while content/content-status.json records qa_passed) to the smallest accurate status line, and added a compact "Source-freshness note (2026-06-15)" section after the "Source-freshness note (2026-06-12)" section. The note is evergreen and source-safe — verify pricing/plans/model availability at the official source, choose by workflow fit over headline specs, and keep a human in the loop on any output. Evidence: a 2026-06-15 KST Hermes recheck found Anthropic's official surfaces reachable (HTTP 200) and stable — https://claude.com/pricing (final URL https://claude.com/pricing, title "Plans & Pricing | Claude by Anthropic", 200000-byte body-sample SHA-256 fbde60e5…, plan-tier names Free/Pro/Max visible), the legacy https://www.anthropic.com/pricing resolving to the same page, https://claude.com/product/overview (HTTP 200), and https://www.anthropic.com/claude-code now resolving to https://claude.com/product/claude-code (HTTP 200) — recorded as src-anthropic-claude-pricing-2026-06-15 (access_status = ok) and added to the claude sources_used. Reachability/title/plan-name evidence only: no exact price, quota, benchmark, ranking, speed, model-availability, or superiority claim was added, and the 2026-05-22 page-body read remains the source of every plan/price quoted here. No Gumroad/UTM/affiliate/sponsored/coupon link added; data/tools.json and last_verified_at unchanged; still qa_passed, still 1 of the 18 tool pages.