New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Rico Bookmark Manager

by ricocc in ricocc/rico-bookmark-manager

Analyzes, deduplicates, reorganizes and exports browser bookmarks from an exported HTML or JSON file, and can build a local bookmark navigation site.

MITAuto-check passedKnowledge Management

Install Rico Bookmark Manager

skills CLI
$ npx skills add ricocc/rico-bookmark-manager --skill rico-bookmark-manager -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install ricocc/rico-bookmark-manager rico-bookmark-manager --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
rico-bookmark-manager
GitHub stars
139
Token cost
~2.5k tokens
SKILL.md length
1,045 words
Files
45 (incl. scripts, references, assets)
Skills in repo
1
Repo updated
First seen
Licence
MIT

At a glance

Analyzes, deduplicates, reorganizes and exports browser bookmarks from an exported HTML or JSON file, and can build a local bookmark navigation site.

  • Works in 5 steps: Clarify the user's goal: analyze,… → Identify the safest available input… → Convert any non-standard source into a… → …
  • Cleaning up and deduplicating an exported bookmarks file
  • SKILL.md covers First Principles Workflow, Decision Flow, Input Acquisition Strategy and Agent Execution Policy, plus 4 more sections
  • Runs JavaScript scripts from its folder; calls python

What it does

A bundled Python command-line tool, `scripts/rico_bookmarks_manager.py`, does the work on browser-exported Netscape HTML files and existing `bookmarks.json` data. It supports structure and duplicate analysis, dead-link checks, search, reclassification, incremental updates into existing data, exports, and a generated local HTML navigation site with built-in themes or a supplied `DESIGN.md`.

The agent first clarifies the goal and picks the safest input, preferring a `bookmarks.html` file you provide over any access to browser profiles. It reads the matching reference file for the task, such as the CLI, workflow, taxonomy, schema or design notes, runs the smallest command that does the job, and reports what data was read, which files were written and whether any browser profile or automation was touched. Vague mentions of bookmarks do not trigger it; the word rico or a clear management action is needed.

When your agent uses it

  • Cleaning up and deduplicating an exported bookmarks file
  • Finding dead links in a large bookmark collection
  • Regrouping bookmarks into a new category structure
  • Building a local navigation page from browser bookmarks

Example prompts

  • “Rico, check bookmarks.html for duplicates and dead links and give me a report.”
  • “Organize my exported bookmarks into categories and export the result as bookmarks.json.”
  • “Generate a local bookmark navigation site from bookmarks.html using one of the built-in themes.”

Requirements

  • Python to run the bundled CLI
  • An exported `bookmarks.html` or an existing `bookmarks.json`

Workflow steps

5 steps, taken from the first numbered list in SKILL.md.

  1. Clarify the user's goal: analyze, organize, deduplicate, check links, update incrementally, export, generate the navigation site, or…
  2. Identify the safest available input source: user-provided bookmarks.html, existing bookmarks.json, agent-visible files, or an authorized…
  3. Convert any non-standard source into a standard input file before invoking the CLI whenever practical.
  4. Run the smallest command that satisfies the request. Prefer deterministic CLI behavior over ad hoc parsing.
  5. Report what data was read, what files were written, and whether any browser profile, browser automation, or current-tab data was accessed.

What it can do on your machine

Read from SKILL.md and the folder at commit 90c9514. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Ships 1 file in scripts/ (JavaScript, from the files we listed), which the agent can run.

    Shell commands in SKILL.md call:

    • python

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Rico Bookmark Manager loads about 2.5k tokens when it runs, and up to ~8.1k if it reads all its reference files. Until then it costs about 183 tokens; SKILL.md has 1,045 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~183
When it runs · the whole SKILL.md, loaded when a task matches
~2.5k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~8.1k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

The full file from ricocc/rico-bookmark-manager at commit 90c9514, republished under its MIT licence (© ricocc). 1,045 words, ~2,472 tokens.

Download SKILL.mdSave it as .claude/skills/rico-bookmark-manager/SKILL.md (or your agent's skills folder). This skill also uses 44 other files; get the full folder from GitHub.
name
rico-bookmark-manager
description
Browser bookmark management and organization: analyze, organize, deduplicate, check dead links, search, export, incrementally update, acquire browser bookmark data, or generate a local bookmark navigation site. Triggers in two cases, in any language: (1) the user mentions "rico" together with bookmark intent (rico is the entry keyword for the rico skill series); or (2) the user expresses a clear, specific browser-bookmark management task using concrete actions, such as organizing, deduplicating, checking dead links, exporting, incrementally updating, or building a navigation site from browser bookmarks. Do not trigger on vague or passing mentions of "bookmark/bookmarks" with no clear management action.

Rico Bookmarks Manager

Use this skill to analyze, organize, update, and export browser bookmarks. The bundled CLI has stable file-based interfaces for browser-exported Netscape HTML files and existing bookmarks.json data. An AI/Agent may help acquire those input files from other sources when the user authorizes it and the environment supports it.

First Principles Workflow

  1. Clarify the user's goal: analyze, organize, deduplicate, check links, update incrementally, export, generate the navigation site, or inspect themes.
  2. Identify the safest available input source: user-provided bookmarks.html, existing bookmarks.json, agent-visible files, or an authorized browser-data acquisition path.
  3. Convert any non-standard source into a standard input file before invoking the CLI whenever practical.
  4. Run the smallest command that satisfies the request. Prefer deterministic CLI behavior over ad hoc parsing.
  5. Report what data was read, what files were written, and whether any browser profile, browser automation, or current-tab data was accessed.

Decision Flow

  1. Confirm the request either includes rico with bookmark intent, or expresses a clear, specific browser-bookmark management task (organizing, deduplicating, checking dead links, exporting, updating, or generating a navigation site). If it is only a vague or passing mention of bookmarks with no clear management action, do not use this skill implicitly.
  2. Resolve the input source using the Input Acquisition Strategy below.
  3. Identify the task:
    • Structure, duplicate, dead-link, search, or distribution analysis: read references/cli.md.
    • Organizing or reclassifying bookmarks: read references/workflow.md and references/taxonomy.md.
    • Incremental update into existing data: read references/schema.md and references/workflow.md.
    • Generating the local HTML bookmark navigation site: read references/manager-spec.md.
    • Selecting a built-in navigation theme or applying a provided DESIGN.md: read references/design-md.md.
  4. Prefer the bundled script scripts/rico_bookmarks_manager.py for deterministic parsing, analysis, exports, and navigation-site generation.
  5. For any generated user-facing surface, determine one display language before writing final output.

Input Acquisition Strategy

  • Prefer user-provided bookmarks.html for full runs, analysis, organization, and incremental update inputs.
  • Prefer existing bookmarks.json for rebuilding the navigation site, exporting Markdown, exporting browser HTML, or merging a new export.
  • If the user asks to get bookmarks from Chrome, Edge, Firefox, or another browser, inspect what the environment can safely access and explain the path you will use.
  • The CLI public interface accepts standard files (--input, --data, --existing). If an agent obtains browser data through profile access, browser automation, or another tool, first convert or copy it into a standard intermediate file before calling the CLI whenever practical.
  • Accessing browser profile files, automating a browser, or reading current open tabs requires explicit user intent and permission. Do not infer permission from a generic request like "整理书签" if browser data access is needed.
  • If direct acquisition is unavailable, ask the user to export bookmarks from the browser and provide the HTML file.

Agent Execution Policy

  • Do not bind behavior to a specific agent product. Any environment that can read files, run Python, and write outputs can execute the core workflow.
  • When a capable agent runtime offers browser automation or local file access, use it only as an input-acquisition layer; keep the processing pipeline centered on the bundled CLI and standard data files.
  • Avoid promising that every agent can read local browser data. Capabilities depend on runtime permissions, OS access, browser state, and user approval.
  • When reporting results, distinguish between CLI capabilities and external agent capabilities.
Show full SKILL.md (506 more words)Show less

Display Language Policy

  • If the user explicitly names a display language, use that language.
  • If the user does not specify a language, follow the main language of the current user request.
  • If the display language is unclear, mixed, conflicts with the bookmark data language, or the user names an audience without a clear language, ask which language to use before generating the navigation site.
  • Keep all template UI copy in a single language: headings, buttons, placeholders, aria labels, empty states, detail modal labels, export menu labels, and in-site Markdown/browser-export labels should not mix languages unless the user explicitly asks for bilingual output.
  • For generated Markdown reports or summaries delivered to the user, localize report headings and explanatory prose to the same display language when needed, while preserving bookmark data values.
  • Bookmark titles, URLs, domains, category names, tags, and source folder names are user data. Preserve them by default instead of translating them.
  • The bundled CLI has no --language option. The bundled navigation template (assets/bookmark-manager-template/index.html and app.js) is hardcoded in Chinese (zh) by default. When the target display language differs, do not patch strings ad hoc: read assets/bookmark-manager-template/i18n-strings.json (a multilingual reference library covering every UI string in zh/en/ja/ko/es/fr), then for each key replace the template's current value with the target-language value. The where and current-zh value in each entry locate the string. Keep bookmark data intact, and keep the brand Rico / Bookmarks untranslated. If a needed language is missing from the library, add it to the JSON first, then apply. After localizing the template, run manager (or all) to generate the site.

Required User Checkpoints

Before organizing, ask only for decisions that are not discoverable from files:

  • Which input source to use if multiple are available: exported HTML, existing JSON, or an authorized agent-created intermediate file.
  • Whether to output by original bookmark folders, optimized Rico taxonomy, or hybrid.
  • Whether to use one, two, or three classification levels.
  • Whether to generate all outputs or only browser import HTML, Markdown, navigation site, reports, or data.
  • Whether this is a full run or an incremental update that should preserve existing classifications.
  • Which navigation theme to use when generating HTML: ease by default, or one of the five built-in themes kami, ease, minimal-mono, retro-blue, ui, or a custom --design file. When offering choices, show all five built-in themes.
  • Which display language to use, but only when the policy above cannot infer it confidently.

Safety Defaults

  • Do not edit Chrome/Edge/Firefox profile files directly.
  • Do not access browser profile files, automate a browser, or read current tabs unless the user clearly wants that path and the environment permits it.
  • Do not delete bookmarks automatically.
  • Do not overwrite human-edited classifications during incremental updates.
  • Treat navigation-site detail edits as browser-local staging by default. They do not write bookmarks.json; tell the user to export the adjusted JSON when they want a replacement data file.
  • Put suspicious, duplicate, low-confidence, and dead-link items into reports for review.
  • Use --dry-run before incremental writes when the user is checking behavior.
  • Back up existing generated data before overwriting it.

Quick Commands

bash
python scripts/rico_bookmarks_manager.py analyze --input bookmarks.html --output rico-bookmarks
python scripts/rico_bookmarks_manager.py organize --input bookmarks.html --output rico-bookmarks --mode optimized --levels 2
python scripts/rico_bookmarks_manager.py export-html --data rico-bookmarks/data/bookmarks.json --output rico-bookmarks
python scripts/rico_bookmarks_manager.py export-md --data rico-bookmarks/data/bookmarks.json --output rico-bookmarks
python scripts/rico_bookmarks_manager.py themes
python scripts/rico_bookmarks_manager.py manager --data rico-bookmarks/data/bookmarks.json --theme kami --output rico-bookmarks/site
python scripts/rico_bookmarks_manager.py manager --data rico-bookmarks/data/bookmarks.json --theme ease --output rico-bookmarks/site
python scripts/rico_bookmarks_manager.py manager --data rico-bookmarks/data/bookmarks.json --theme minimal-mono --output rico-bookmarks/site
python scripts/rico_bookmarks_manager.py manager --data rico-bookmarks/data/bookmarks.json --theme retro-blue --output rico-bookmarks/site
python scripts/rico_bookmarks_manager.py manager --data rico-bookmarks/data/bookmarks.json --theme ui --output rico-bookmarks/site
python scripts/rico_bookmarks_manager.py manager --data rico-bookmarks/data/bookmarks.json --design DESIGN.md --output rico-bookmarks/site
python scripts/rico_bookmarks_manager.py update --input new.html --existing rico-bookmarks/data/bookmarks.json --output rico-bookmarks --dry-run
python scripts/rico_bookmarks_manager.py all --input bookmarks.html --output rico-bookmarks --theme kami

© ricocc, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 44 other files (scripts, references, assets) in the repository root of ricocc/rico-bookmark-manager.

  • SKILL.md
  • .gitignore
  • LICENSE
  • README.md
  • README.zh.md
  • agents/openai.yaml
  • assets/bookmark-manager-template/app.js
  • assets/bookmark-manager-template/i18n-strings.json
  • assets/bookmark-manager-template/index.html
  • assets/bookmark-manager-template/logo.jpg
  • assets/bookmark-manager-template/styles.css
  • assets/themes/ease/DESIGN.md
  • assets/themes/ease/theme.css
  • assets/themes/ease/theme.json
  • assets/themes/kami/DESIGN.md
  • … and 30 more

Open the folder on GitHubat commit 90c9514

Compare with similar skills

Rico Bookmark Manager next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Rico Bookmark Manager compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Rico Bookmark Manager this skillricocc/rico-bookmark-manager139—~2.5kAutomated safety check: PassMIT
Defuddlekepano/obsidian-skills49k11 repos~208Automated safety check: PassMIT
URL to Markdown FetcherJimLiu/baoyu-skills27k1 repos~2.1kAutomated safety check: PassMIT
Xiaohongshu Post Extractorchenxiachan/xhs-claude-skills421—~1.5kAutomated safety check: WarnMIT
Baoyu URL To Markdownsdyckjq-lab/llm-wiki-skill2.5k2 repos~3.2kAutomated safety check: PassNone
X to Markdown ConverterJimLiu/baoyu-skills27k4 repos~1.8kAutomated safety check: WarnMIT

Similar skills

  • Defuddle

    kepano/obsidian-skills

    Uses the Defuddle CLI to pull clean, readable Markdown, JSON or metadata from web pages, stripping navigation, ads and clutter to save tokens.

    49k GitHub starsUsed in 11 repos~208 tokens
    Knowledge ManagementAuto-check passed

  • URL to Markdown Fetcher

    JimLiu/baoyu-skills

    Fetches a web page, X post, YouTube transcript or Hacker News thread through a Chrome-driven CLI and saves it as clean markdown.

    27k GitHub starsUsed in 1 repo~2.1k tokens
    Knowledge ManagementAuto-check passed

  • Xiaohongshu Post Extractor

    chenxiachan/xhs-claude-skills

    Pulls a Xiaohongshu post's text, image text and video subtitles or transcript into a Markdown note saved in an Obsidian folder, in Chinese.

    421 GitHub stars~1.5k tokensUpdated 1 mo ago
    Knowledge ManagementAuto-check: warnings

  • Baoyu URL To Markdown

    sdyckjq-lab/llm-wiki-skill

    Fetch any URL and convert to markdown using Chrome CDP. An agent skill from sdyckjq-lab/llm-wiki-skill.

    2.5k GitHub starsUsed in 2 repos~3.2k tokens
    Knowledge ManagementAuto-check passed

  • X to Markdown Converter

    JimLiu/baoyu-skills

    Saves tweets, threads and X Articles as Markdown files with YAML front matter, using an unofficial API that asks for your consent first.

    27k GitHub starsUsed in 4 repos~1.8k tokens
    Knowledge ManagementAuto-check: warnings

  • Icm Architect

    RinDig/icm-architect

    Design any process, idea, problem, or body of knowledge into an ICM (Interpretable Context Methodology) workspace — folder structure as agent architecture — or restructure an existing folder, repo…

    1.8k GitHub stars~3.2k tokensUpdated 1 mo ago
    Knowledge ManagementAuto-check passed

Categories

Questions about Rico Bookmark Manager

What does Rico Bookmark Manager do?

Analyzes, deduplicates, reorganizes and exports browser bookmarks from an exported HTML or JSON file, and can build a local bookmark navigation site. json` data.md`.

When should I use Rico Bookmark Manager?

Rico Bookmark Manager fits situations like: cleaning up and deduplicating an exported bookmarks file; finding dead links in a large bookmark collection; regrouping bookmarks into a new category structure; building a local navigation page from browser bookmarks.

How do I install Rico Bookmark Manager in Claude Code?

Run `npx skills add ricocc/rico-bookmark-manager --skill rico-bookmark-manager -a claude-code`. Or copy the skill folder (the ricocc/rico-bookmark-manager repository) into .claude/skills/rico-bookmark-manager in your project. Claude Code loads it when a task matches its description.

How do I install Rico Bookmark Manager in Codex?

Run `npx skills add ricocc/rico-bookmark-manager --skill rico-bookmark-manager -a codex`. Or copy the skill folder (the ricocc/rico-bookmark-manager repository) into .agents/skills/rico-bookmark-manager in your project. Codex loads it when a task matches its description.

Can I use Rico Bookmark Manager in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add ricocc/rico-bookmark-manager --skill rico-bookmark-manager -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/rico-bookmark-manager, .gemini/skills/rico-bookmark-manager, .github/skills/rico-bookmark-manager and .opencode/skills/rico-bookmark-manager in your project.

What does Rico Bookmark Manager need to run?

Going by SKILL.md and its folder, Rico Bookmark Manager needs JavaScript for the scripts in its folder and the command-line tools its instructions call (python). Our summary lists: Python to run the bundled CLI; An exported `bookmarks.html` or an existing `bookmarks.json`.

Does Rico Bookmark Manager access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Rico Bookmark Manager safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.

What licence does Rico Bookmark Manager use?

Rico Bookmark Manager is published under the MIT licence (from the LICENSE file in the skill folder). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Rico Bookmark Manager use?

About 2.5k tokens (SKILL.md is roughly 9.9k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 5.7k tokens, read only when the agent opens those files.

What are the alternatives to Rico Bookmark Manager?

Skills that share tags, products or a category with Rico Bookmark Manager: Defuddle (kepano/obsidian-skills, 49k stars), URL to Markdown Fetcher (JimLiu/baoyu-skills, 27k stars), Xiaohongshu Post Extractor (chenxiachan/xhs-claude-skills, 421 stars) and Baoyu URL To Markdown (sdyckjq-lab/llm-wiki-skill, 2.5k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Rico Bookmark Manager?

ricocc (a GitHub user) maintains it in ricocc/rico-bookmark-manager, which has 139 GitHub stars. The repository was last updated on September 10, 2026.

Source: ricocc/rico-bookmark-manager on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.