yuno-proofreader 0.1.0

data/lib/yuno/proofreader/version.rb

@@ -0,0 +1,7 @@
+# frozen_string_literal: true
+
+module Yuno
+  module Proofreader
+    VERSION = "0.1.0"
+  end
+end

data/lib/yuno/proofreader.rb

@@ -0,0 +1,101 @@
+# frozen_string_literal: true
+
+require_relative "proofreader/version"
+require_relative "proofreader/errors"
+require_relative "proofreader/config"
+require_relative "proofreader/log"
+require_relative "proofreader/site"
+require_relative "proofreader/env_file"
+require_relative "proofreader/http/client"
+require_relative "proofreader/http/robots"
+require_relative "proofreader/sitemap/parser"
+require_relative "proofreader/sitemap/aligner"
+require_relative "proofreader/sitemap/mismatch"
+require_relative "proofreader/sitemap/pairer"
+require_relative "proofreader/sitemap/loader"
+require_relative "proofreader/sitemap/report"
+require_relative "proofreader/sitemap/importer"
+require_relative "proofreader/pages/trimmer"
+require_relative "proofreader/pages/snapshot"
+require_relative "proofreader/pages/fetcher"
+require_relative "proofreader/pages/report"
+require_relative "proofreader/store/schema"
+require_relative "proofreader/store/records"
+require_relative "proofreader/store/cells"
+require_relative "proofreader/store/memory"
+require_relative "proofreader/store/sheets"
+require_relative "proofreader/store/tab_setup"
+require_relative "proofreader/context"
+require_relative "proofreader/site_reader"
+require_relative "proofreader/findings/text"
+require_relative "proofreader/findings/finding"
+require_relative "proofreader/findings/outline"
+require_relative "proofreader/findings/source"
+require_relative "proofreader/findings/matching"
+require_relative "proofreader/findings/merger"
+require_relative "proofreader/findings/plan"
+require_relative "proofreader/findings/reconciler"
+require_relative "proofreader/analysis/links"
+require_relative "proofreader/analysis/structured_data"
+require_relative "proofreader/analysis/extractor"
+require_relative "proofreader/analysis/url_tree"
+require_relative "proofreader/analysis/crawler"
+require_relative "proofreader/analysis/settings"
+require_relative "proofreader/analysis/list_order"
+require_relative "proofreader/analysis/adjustments"
+require_relative "proofreader/analysis/weights"
+require_relative "proofreader/analysis/graph"
+require_relative "proofreader/analysis/signals"
+require_relative "proofreader/analysis/reasons"
+require_relative "proofreader/analysis/entries"
+require_relative "proofreader/analysis/judgments"
+require_relative "proofreader/analysis/judge_prompt"
+require_relative "proofreader/analysis/judge"
+require_relative "proofreader/analysis/scorer"
+require_relative "proofreader/analysis/runner"
+require_relative "proofreader/analysis/report"
+require_relative "proofreader/llm/errors"
+require_relative "proofreader/llm/types"
+require_relative "proofreader/llm/json_text"
+require_relative "proofreader/llm/http_transport"
+require_relative "proofreader/llm/provider"
+require_relative "proofreader/llm/providers"
+require_relative "proofreader/llm/providers/gemini"
+require_relative "proofreader/llm/defaults"
+require_relative "proofreader/llm/client"
+require_relative "proofreader/llm/factory"
+require_relative "proofreader/llm/usage"
+require_relative "proofreader/audit/steps"
+require_relative "proofreader/audit/instructions"
+require_relative "proofreader/audit/prompt"
+require_relative "proofreader/audit/answers"
+require_relative "proofreader/audit/attempt"
+require_relative "proofreader/audit/runner"
+require_relative "proofreader/audit/report"
+require_relative "proofreader/audit/log_fields"
+require_relative "proofreader/audit/cadence"
+require_relative "proofreader/audit/agenda"
+require_relative "proofreader/audit/batch"
+require_relative "proofreader/audit/agenda_report"
+require_relative "proofreader/audit/batch_report"
+require_relative "proofreader/cli/audit_commands"
+require_relative "proofreader/cli/command_log"
+require_relative "proofreader/cli/model_commands"
+require_relative "proofreader/cli/sheet_commands"
+require_relative "proofreader/cli/analysis_commands"
+require_relative "proofreader/settings_file"
+require_relative "proofreader/init/parsing"
+require_relative "proofreader/init/detector"
+require_relative "proofreader/init/ask"
+require_relative "proofreader/init/web"
+require_relative "proofreader/init/interview"
+require_relative "proofreader/cli/flags"
+require_relative "proofreader/cli/init_command"
+require_relative "proofreader/cli/queue_commands"
+require_relative "proofreader/cli"
+
+module Yuno
+  # AI-powered website proofreader: reads a site, audits its pages and lists suggested fixes.
+  module Proofreader
+  end
+end

data/proofreader.example.yml

@@ -0,0 +1,134 @@
+# Copy this file to proofreader.yml and edit it. `yuno-proofreader settings` shows what is still to fill in.
+
+# The address of the site, without a path.
+website: https://example.com
+
+# One entry per language. The first one is the default language: pages are named after its addresses,
+# and default_language_only_paths refers to it.
+#
+#   code    The language code. Always put it in quotes: without them YAML reads a bare no (the code for
+#           Norwegian) as false.
+#   prefix  What the web address of a page in this language starts with, after the domain:
+#             ""     nothing, as in example.com/about
+#             "/en"  as in example.com/en/about
+#   notes   Optional. How this language should read, such as its regional variant or whether it addresses
+#           readers formally. It is sent to the AI model, see "About the site" below.
+languages:
+  - code: "de"      # German
+    prefix: ""
+    notes: "[FILL IN] The variant and form of address of this language, for example: formal address (Sie)."
+  - code: "es"      # Spanish
+    prefix: "/es"
+    notes: "[FILL IN] For example: Latin American Spanish, formal address (usted)."
+  - code: "en"      # English
+    prefix: "/en"
+    notes: "[FILL IN] For example: American English."
+
+# A site where every language has a prefix, and English is the default:
+# languages:
+#   - code: "en"
+#     prefix: "/en"
+#   - code: "de"
+#     prefix: "/de"
+
+# Where your sitemap is. Leave this out and the tool looks in this order:
+#   1. the Sitemap lines in https://example.com/robots.txt (every one of them is read)
+#   2. https://example.com/sitemap.xml
+# Set it if your sitemap lives somewhere else. It takes one address, and when it is set robots.txt is not read.
+# sitemap_url: https://example.com/sitemap.xml
+
+# Pages that exist only in the default language, written as they appear in the address.
+# A path also covers everything below it.
+# default_language_only_paths: ["/news"]
+
+# Pages you do not want audited, in any language. A path also covers everything below it.
+# skip_paths: ["/thank-you", "/es/gracias"]
+
+# ---- About the site ----------------------------------------------------------------------------
+# Everything below is sent to the AI model together with each page, so keep it short and free of secrets.
+# Limits: 500 characters per text, 200 per entry of a list or note of a language, 50 entries per list.
+#
+# Every value here starts with [FILL IN]. The commands that send text to the model (audit, prompt, and
+# analyze unless you pass --no-ai) refuse to run while that text is still there: replace the whole value
+# with your own text, or delete the setting. Only business is required.
+
+# What the site is and what it offers, in a sentence or two.
+business: "[FILL IN] What the site is and what it offers, in a sentence or two."
+
+# Who reads the site.
+audience: "[FILL IN] Who reads the site (end user)."
+
+# How the text should sound.
+tone: "[FILL IN] How the text should sound (formal, informal, etc)."
+
+# Writing rules, one per entry.
+rules:
+  - "[FILL IN] One writing rule of your own, such as how prices or dates are written."
+  - "[FILL IN] Another writing rule, or delete this line."
+
+# Words that are correct as written and must never be flagged, one per entry: your brand name,
+# place names, product names, local words.
+ignore_terms:
+  - "[FILL IN] One term that is correct as written, such as your brand name."
+
+# Differences between language versions that are intended, one per entry.
+intentional_differences:
+  - "[FILL IN] One difference between language versions that is intended, such as a section that exists in one language only."
+
+# ---- The AI models ----------------------------------------------------------------------------
+# Everything here is optional.
+#
+#   provider       Which family of models to use. Leave it out while only one is installed.
+#   models         The models to use, best first. When one is overloaded, over its limit or not available to
+#                  your key, the next one is used. The tool has a built-in list to fall back on, but model
+#                  names change often, so set your own. `yuno-proofreader models` shows which ones your key
+#                  can use. The names below are only examples.
+#   review_models  Optional list for the independent second review, which prefers a different model from the
+#                  first audit. Leave it out to use `models`. With only one model there is nothing for the
+#                  review to be independent of.
+#
+# provider: gemini
+# models:
+#   - gemini-3.8-flash
+#   - gemini-3.7-flash
+# review_models:
+#   - gemini-3.6-flash
+
+# ---- The analysis of the site (optional) -------------------------------------------------------
+# Used by `yuno-proofreader analyze`, which ranks the pages of the site by importance. Everything here is
+# optional. The values below are examples, not placeholders: if you remove the # signs without changing
+# them, they are applied and sent to the model as they stand.
+#
+#   importance_hint  What the model should know about your business that it cannot tell from the pages,
+#                    such as which offers matter most. It is sent to the AI model.
+#   priority_paths   Points added to the importance of the pages at or below a path, from -50 to 50.
+#                    The most specific path wins. Use negative points for pages that matter less.
+#   hubs             Index pages to read even if nothing else leads to them. The tool already finds the
+#                    pages that have other pages under them.
+#   hub_order        editorial: the order of your lists says what matters most. neutral: it says nothing.
+#                    auto (the default): editorial, unless a list is in alphabetical order.
+#   max_pages        How many pages the analysis reads at most, from 1 to 500 (default 80).
+#
+# analysis:
+#   importance_hint: "What we sell ourselves matters more than the information pages."
+#   priority_paths:
+#     /products: 10
+#     /careers: -20
+#   hubs: ["/blog"]
+#   hub_order: auto
+#   max_pages: 80
+
+# ---- How often pages are audited again (optional) ----------------------------------------------
+# Used by `yuno-proofreader queue` and `audit --next`, which pick the pages that are due. The tool schedules
+# nothing itself: run `audit --next` from cron, a Kubernetes CronJob or any other scheduler.
+# The values below are the defaults, so removing the # signs without changing them changes nothing.
+#
+#   intervals      Days between audits for each tier of the analysis (1 is the most important). Tiers you
+#                  leave out keep their default.
+#   retry_days     Days before a page whose last audit was incomplete is tried again.
+#   default_tier   The tier a page counts as until the analysis has ranked it.
+#
+# cadence:
+#   intervals: { 1: 30, 2: 60, 3: 120, 4: 240 }
+#   retry_days: 7
+#   default_tier: 3

metadata

@@ -0,0 +1,194 @@
+--- !ruby/object:Gem::Specification
+name: yuno-proofreader
+version: !ruby/object:Gem::Version
+  version: 0.1.0
+platform: ruby
+authors:
+- Sebastian Alsen
+bindir: exe
+cert_chain: []
+date: 1980-01-02 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: google-apis-sheets_v4
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0.49'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0.49'
+- !ruby/object:Gem::Dependency
+  name: googleauth
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.17'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.17'
+- !ruby/object:Gem::Dependency
+  name: nokogiri
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.19'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.19'
+description: Reviews your public web pages with an AI model (Gemini so far) for typos,
+  grammar, HTML structure, SEO and differences between language versions, ranks your
+  pages by importance and lists the suggested fixes in a Google Sheet for you to review.
+executables:
+- yuno-proofreader
+extensions: []
+extra_rdoc_files: []
+files:
+- CHANGELOG.md
+- LICENSE.txt
+- README.md
+- exe/yuno-proofreader
+- lib/yuno/proofreader.rb
+- lib/yuno/proofreader/analysis/adjustments.rb
+- lib/yuno/proofreader/analysis/crawler.rb
+- lib/yuno/proofreader/analysis/entries.rb
+- lib/yuno/proofreader/analysis/extractor.rb
+- lib/yuno/proofreader/analysis/graph.rb
+- lib/yuno/proofreader/analysis/judge.rb
+- lib/yuno/proofreader/analysis/judge.yml
+- lib/yuno/proofreader/analysis/judge_prompt.rb
+- lib/yuno/proofreader/analysis/judgments.rb
+- lib/yuno/proofreader/analysis/links.rb
+- lib/yuno/proofreader/analysis/list_order.rb
+- lib/yuno/proofreader/analysis/reasons.rb
+- lib/yuno/proofreader/analysis/report.rb
+- lib/yuno/proofreader/analysis/runner.rb
+- lib/yuno/proofreader/analysis/scorer.rb
+- lib/yuno/proofreader/analysis/settings.rb
+- lib/yuno/proofreader/analysis/signals.rb
+- lib/yuno/proofreader/analysis/structured_data.rb
+- lib/yuno/proofreader/analysis/url_tree.rb
+- lib/yuno/proofreader/analysis/weights.rb
+- lib/yuno/proofreader/analysis/weights.yml
+- lib/yuno/proofreader/audit/agenda.rb
+- lib/yuno/proofreader/audit/agenda_report.rb
+- lib/yuno/proofreader/audit/answers.rb
+- lib/yuno/proofreader/audit/attempt.rb
+- lib/yuno/proofreader/audit/batch.rb
+- lib/yuno/proofreader/audit/batch_report.rb
+- lib/yuno/proofreader/audit/cadence.rb
+- lib/yuno/proofreader/audit/instructions.rb
+- lib/yuno/proofreader/audit/log_fields.rb
+- lib/yuno/proofreader/audit/prompt.rb
+- lib/yuno/proofreader/audit/prompts/instructions.yml
+- lib/yuno/proofreader/audit/prompts/steps.yml
+- lib/yuno/proofreader/audit/report.rb
+- lib/yuno/proofreader/audit/runner.rb
+- lib/yuno/proofreader/audit/steps.rb
+- lib/yuno/proofreader/cli.rb
+- lib/yuno/proofreader/cli/analysis_commands.rb
+- lib/yuno/proofreader/cli/audit_commands.rb
+- lib/yuno/proofreader/cli/command_log.rb
+- lib/yuno/proofreader/cli/flags.rb
+- lib/yuno/proofreader/cli/init_command.rb
+- lib/yuno/proofreader/cli/model_commands.rb
+- lib/yuno/proofreader/cli/queue_commands.rb
+- lib/yuno/proofreader/cli/sheet_commands.rb
+- lib/yuno/proofreader/config.rb
+- lib/yuno/proofreader/context.rb
+- lib/yuno/proofreader/env_file.rb
+- lib/yuno/proofreader/errors.rb
+- lib/yuno/proofreader/findings/finding.rb
+- lib/yuno/proofreader/findings/matching.rb
+- lib/yuno/proofreader/findings/merger.rb
+- lib/yuno/proofreader/findings/outline.rb
+- lib/yuno/proofreader/findings/plan.rb
+- lib/yuno/proofreader/findings/reconciler.rb
+- lib/yuno/proofreader/findings/source.rb
+- lib/yuno/proofreader/findings/text.rb
+- lib/yuno/proofreader/http/client.rb
+- lib/yuno/proofreader/http/net_http_transport.rb
+- lib/yuno/proofreader/http/robots.rb
+- lib/yuno/proofreader/init/ask.rb
+- lib/yuno/proofreader/init/detector.rb
+- lib/yuno/proofreader/init/interview.rb
+- lib/yuno/proofreader/init/parsing.rb
+- lib/yuno/proofreader/init/web.rb
+- lib/yuno/proofreader/llm/client.rb
+- lib/yuno/proofreader/llm/defaults.rb
+- lib/yuno/proofreader/llm/defaults.yml
+- lib/yuno/proofreader/llm/errors.rb
+- lib/yuno/proofreader/llm/factory.rb
+- lib/yuno/proofreader/llm/http_transport.rb
+- lib/yuno/proofreader/llm/json_text.rb
+- lib/yuno/proofreader/llm/model_commands.rb
+- lib/yuno/proofreader/llm/provider.rb
+- lib/yuno/proofreader/llm/providers.rb
+- lib/yuno/proofreader/llm/providers/gemini.rb
+- lib/yuno/proofreader/llm/types.rb
+- lib/yuno/proofreader/llm/usage.rb
+- lib/yuno/proofreader/log.rb
+- lib/yuno/proofreader/pages/fetcher.rb
+- lib/yuno/proofreader/pages/report.rb
+- lib/yuno/proofreader/pages/snapshot.rb
+- lib/yuno/proofreader/pages/trimmer.rb
+- lib/yuno/proofreader/settings_file.rb
+- lib/yuno/proofreader/site.rb
+- lib/yuno/proofreader/site_reader.rb
+- lib/yuno/proofreader/sitemap/aligner.rb
+- lib/yuno/proofreader/sitemap/importer.rb
+- lib/yuno/proofreader/sitemap/loader.rb
+- lib/yuno/proofreader/sitemap/mismatch.rb
+- lib/yuno/proofreader/sitemap/pairer.rb
+- lib/yuno/proofreader/sitemap/parser.rb
+- lib/yuno/proofreader/sitemap/report.rb
+- lib/yuno/proofreader/store/cells.rb
+- lib/yuno/proofreader/store/google_gateway.rb
+- lib/yuno/proofreader/store/memory.rb
+- lib/yuno/proofreader/store/records.rb
+- lib/yuno/proofreader/store/schema.rb
+- lib/yuno/proofreader/store/sheets.rb
+- lib/yuno/proofreader/store/tab_setup.rb
+- lib/yuno/proofreader/version.rb
+- proofreader.example.yml
+homepage: https://gitlab.com/alsen-ca/yuno-proofreader
+licenses:
+- MIT
+metadata:
+  source_code_uri: https://gitlab.com/alsen-ca/yuno-proofreader
+  bug_tracker_uri: https://gitlab.com/alsen-ca/yuno-proofreader/-/issues
+  changelog_uri: https://gitlab.com/alsen-ca/yuno-proofreader/-/blob/main/CHANGELOG.md
+  rubygems_mfa_required: 'true'
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '4.0'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubygems_version: 4.0.20
+specification_version: 4
+summary: AI-powered website proofreader that collects its suggestions in Google Sheets.
+test_files: []