Tmux
trpc-group/trpc-agent-go
Remote-control tmux sessions for interactive CLIs by sending keystrokes and scraping pane output.
Rewrite messy or AI-generated crawler code into clean, production-ready style that follows the zavod best practices.
Install Refactor Crawler
$ npx skills add opensanctions/opensanctions --skill refactor-crawler -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install opensanctions/opensanctions refactor-crawler --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/opensanctions/opensanctions.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills/refactor-crawler .claude/skills/refactor-crawler && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "refactor-crawler" agent skill from https://github.com/opensanctions/opensanctions/tree/main/.claude/skills/refactor-crawler into .claude/skills/refactor-crawler/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "refactor-crawler", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/opensanctions/opensanctions/tree/main/.claude/skills/refactor-crawlerType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add opensanctions/opensanctions --skill refactor-crawler -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install opensanctions/opensanctions refactor-crawler --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/opensanctions/opensanctions.git skills-src && mkdir -p .agents/skills && cp -r skills-src/.claude/skills/refactor-crawler .agents/skills/refactor-crawler && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "refactor-crawler" agent skill from https://github.com/opensanctions/opensanctions/tree/main/.claude/skills/refactor-crawler into .agents/skills/refactor-crawler/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "refactor-crawler", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add opensanctions/opensanctions --skill refactor-crawler -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install opensanctions/opensanctions refactor-crawler --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/opensanctions/opensanctions.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/.claude/skills/refactor-crawler .cursor/skills/refactor-crawler && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "refactor-crawler" agent skill from https://github.com/opensanctions/opensanctions/tree/main/.claude/skills/refactor-crawler into .cursor/skills/refactor-crawler/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "refactor-crawler", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/opensanctions/opensanctions.git --path .claude/skills/refactor-crawler--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add opensanctions/opensanctions --skill refactor-crawler -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install opensanctions/opensanctions refactor-crawler --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/opensanctions/opensanctions.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/.claude/skills/refactor-crawler .gemini/skills/refactor-crawler && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "refactor-crawler" agent skill from https://github.com/opensanctions/opensanctions/tree/main/.claude/skills/refactor-crawler into .gemini/skills/refactor-crawler/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "refactor-crawler", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install opensanctions/opensanctions refactor-crawlerInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add opensanctions/opensanctions --skill refactor-crawler -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/opensanctions/opensanctions.git skills-src && mkdir -p .github/skills && cp -r skills-src/.claude/skills/refactor-crawler .github/skills/refactor-crawler && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "refactor-crawler" agent skill from https://github.com/opensanctions/opensanctions/tree/main/.claude/skills/refactor-crawler into .github/skills/refactor-crawler/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "refactor-crawler", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add opensanctions/opensanctions --skill refactor-crawler -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install opensanctions/opensanctions refactor-crawler --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/opensanctions/opensanctions.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/.claude/skills/refactor-crawler .opencode/skills/refactor-crawler && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "refactor-crawler" agent skill from https://github.com/opensanctions/opensanctions/tree/main/.claude/skills/refactor-crawler into .opencode/skills/refactor-crawler/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "refactor-crawler", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
refactor-crawler- GitHub stars
- 831
- Token cost
- ~1.1k tokens
- SKILL.md length
- 452 words
- Files
- 1
- Skills in repo
- 11
- Repo updated
- First seen
- Licence
- MIT
At a glance
Rewrite messy or AI-generated crawler code into clean, production-ready style that follows the zavod best practices.
- Works in 5 steps: Read the crawler and its .yml. Note the… → Capture a baseline: run zavod crawl on… → Read the docs for the areas the crawler… → …
- The user asks to clean up
- SKILL.md covers Golden rule: every output…, Prime directive: strict…, The docs and Workflow, plus 1 more section
- Calls ruff and mypy
What it does
Refactor Crawler is an agent skill from opensanctions/opensanctions. Rewrite messy or AI-generated crawler code into clean, production-ready style that follows the zavod best practices. Use when the user asks to clean up, refactor, tidy, or "make production-ready" a crawler, or to bring code in line with best practices.
Its SKILL.md is about 1.1k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.
It sits in Data & Analytics, covering Web scraping and Refactoring. The repository describes itself as: An open database of international sanctions data, persons of interest and politically exposed persons. The licence is MIT.
When your agent uses it
- The user asks to clean up
- Make production-ready a crawler
- Bring code in line with best practices
Example prompts
- “make production-ready”
- “/refactor-crawler”
Requirements
- Pre-approved tools (allowed-tools): Read, Edit, Write, Glob, Grep, Bash
Workflow steps
5 steps, taken from the first numbered list in SKILL.md.
- Read the crawler and its .yml. Note the data source shape (HTML/CSV/XLSX/JSON/API)
- Capture a baseline: run zavod crawl on the unmodified crawler and keep the
- Read the docs for the areas the crawler touches and apply them, most impactful
- Verify (below).
- Summarise: which best-practice areas changed, every output change and its
What it can do on your machine
Read from SKILL.md and the folder at commit 4499adf. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves these tools, so the agent can use them without asking each time:
ReadEditWriteGlobGrepBash
From allowed-tools in the SKILL.md frontmatter.
Runs code
Shell commands in SKILL.md call:
ruffmypy
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
Refactor Crawler loads about 1.1k tokens when it runs. Until then it costs about 67 tokens; SKILL.md has 452 words of instructions outside code blocks.
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: notes
The automated check noted patterns worth knowing about, such as sudo or a known installer.
- NotePre-approves every shell command (allowed-tools: Bash)SKILL.md
allowed-tools: Read, Edit, Write, Glob, Grep, Bash
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); files beside SKILL.md are not scanned.
SKILL.md
The full file from opensanctions/opensanctions at commit 4499adf, republished under its MIT licence (© opensanctions). 452 words, ~1,071 tokens.
Download SKILL.mdSave it as
.claude/skills/refactor-crawler/SKILL.md (or your agent's skills folder).- name
- refactor-crawler
- description
- Rewrite messy or AI-generated crawler code into clean, production-ready style that follows the zavod best practices. Use when the user asks to clean up, refactor, tidy, or "make production-ready" a crawler, or to bring code in line with best practices.
- allowed-tools
- Read, Edit, Write, Glob, Grep, Bash
- argument-hint
- [crawler.py path]
Refactor crawler
Bring an existing crawler in line with current best practices. Typical input is a first-draft or AI-generated crawler that works but is verbose, brittle, or non-idiomatic. Target path: $ARGUMENTS
The rules live in zavod/docs/best_practices/ — this skill is only the procedure for
applying them. When a rule is unclear, read the doc; do not invent a convention.
Golden rule: every output change must be deliberate
Output may change when the refactor brings a methodological gain — e.g. moving to
h.apply_date (warnings + original_value), h.make_address composition, a
type.address lookup dropping placeholders, a tightened selector excluding junk. Name
each such change in your summary and tie it to the method that caused it. Never silently
alter emitted values. Leave name cleaning as it is: changing it means the sequenced,
multi-PR migration to the review system in zavod/docs/extract/names.md, out of scope
for a refactor pass.
Prime directive: strict interpretation
zavod/docs/best_practices/strict_interpretation.md — every source value within the
crawler's scope is handled, explicitly ignored, or raises a signal. A refactored crawler
must be more likely to fail loudly than the draft it replaces, never less: when you
remove a branch, replace it with a guard, not with silence.
The docs
best_practices/strict_interpretation.md— the prime directive: destructive parsing,audit_data, assertions, categorical coveragebest_practices/patterns.md— structure, naming, helpers, constants, logging, pagination, text hygienebest_practices/xpath_and_html.md— typed HTML helpers, selector qualitybest_practices/entity_id.md—make_idvsmake_slug, which fields,key=best_practices/datapatch_lookups.md— replacing inline conditionals with YAML lookupsbest_practices/dates_meta.md—apply_dateand dataset-level date formatsbest_practices/addresses.md—make_address+copy_addressbest_practices/http_operations.md—context.fetch_*, headers, Zytebest_practices/caching.md— what not to cachebest_practices/priorities.md— which properties are worth the effortbest_practices/merge_checklist.md— the pre-merge review checklist
For pure type-annotation errors, defer to the /typechecker-fixes skill.
Show full SKILL.md (181 more words)Show less
Workflow
- Read the crawler and its
.yml. Note the data source shape (HTML/CSV/XLSX/JSON/API) and the existinglookups:,dates:, andhttp:sections. - Capture a baseline: run
zavod crawlon the unmodified crawler and keep the statement counts and a few sample entities to diff against later. - Read the docs for the areas the crawler touches and apply them, most impactful first: structure → selectors/parsing → IDs → dates/addresses → lookups → HTTP/caching → logging/assertions → style nits.
- Verify (below).
- Summarise: which best-practice areas changed, every output change and its methodological justification, and anything deliberately left alone.
Verification
From the repo root:
ruff check --fix --select I <path/to/crawler.py> # import order
cd zavod && mypy --strict <path/to/crawler.py> # enforced by the mypy-datasets hook
zavod crawl <path/to/dataset.yml> # from the repo rootThen confirm:
- New strict-interpretation guards are welcome — but when one fires on today's data, add
the missing lookup mapping rather than handing the warning off. The facility for many
warnings stays in the code;
issues.logis clean (transient network errors excepted). - Diff the output against the baseline run (statement counts, spot-check entities) and account for every difference — each must trace to a named methodological gain.
- The crawler still satisfies its
assertions:in the.yml. - Walk the
merge_checklist.mditems relevant to what you touched.
© opensanctions, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
Just SKILL.md in .claude/skills/refactor-crawler of opensanctions/opensanctions.
Open the folder on GitHubat commit 4499adf
Compare with similar skills
Refactor Crawler 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.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Refactor Crawler this skillopensanctions/opensanctions | 831 | — | ~1.1k | Automated safety check: Notes | MIT | |
| Tmuxtrpc-group/trpc-agent-go | 1.8k | 23 repos | ~868 | Automated safety check: Pass | Apache-2.0 | |
| Ketch1broseidon/ketch | 696 | 1 repos | ~3.9k | Automated safety check: Pass | MIT | |
| Crawl4AI Web Scrapingsmallnest/goclaw | 598 | 1 repos | ~2.5k | Automated safety check: Pass | MIT | |
| Boss Zhipin Scrapereatmoreduck/boss-zhipin-scraper | 1.5k | — | ~2.6k | Automated safety check: Pass | MIT | |
| Axyusukebe/ax | 719 | 1 repos | ~918 | Automated safety check: Pass | MIT |
Similar skills
Ketch
1broseidon/ketch
Research skill for ketch — a fast stateless CLI for web search, OSS code search, curated library docs, page scraping, and site crawling; an optional MCP server exists for operators who want it, but…
Crawl4AI Web Scraping
smallnest/goclaw
Scrapes sites, handles JavaScript-heavy pages and extracts structured data with Crawl4AI, through its crwl CLI or Python SDK, including schema-based extraction without an LLM.
Boss Zhipin Scraper
eatmoreduck/boss-zhipin-scraper
Scrape BOSS直聘 (job listing site) via Chrome CDP. An agent skill from eatmoreduck/boss-zhipin-scraper.
Ax
yusukebe/ax
Use the ax CLI instead of curl + throwaway parsing scripts whenever you fetch a URL, explore an unknown web page, or extract structured data from HTML.
Anakinscraper
Anakin-Inc/anakin
Scrape any website into clean markdown or structured JSON. An agent skill from Anakin-Inc/anakin.
More from opensanctions/opensanctions
All 11 skills in this repoCrawler Pep
opensanctions/opensanctions
Scaffold a new PEP (Politically Exposed Persons) crawler — members of a parliament, legislature, senate, chamber of deputies, cabinet, judiciary, or an asset-declaration register — from a source URL…
Crawler Constants To Yml
opensanctions/opensanctions
Move hardcoded lookup/config constants (gender maps, header dicts, value translations, column-label maps, date formats) out of a crawler and into the dataset .yml — as datapatch lookups wherever…
Dataset Metadata
opensanctions/opensanctions
Bring a dataset .yml's metadata in line with house conventions (title, summary, description, coverage, publisher, maintainer comments).
Legislature Metadata
opensanctions/opensanctions
Refactor the title, description and coverage frequency of a legislature/parliament PEP dataset .yml into the house style.
Name Framework Migration First Step
opensanctions/opensanctions
Migrate ad-hoc name cleaning in a crawler to h.reviewnames (Step 1 of the name framework migration).
Release Datasets
opensanctions/opensanctions
Release one or more datasets by adding them to a topical collection, bumping coverage.start, and verifying.
Related
Categories
Questions about Refactor Crawler
What does Refactor Crawler do?
Rewrite messy or AI-generated crawler code into clean, production-ready style that follows the zavod best practices. Refactor Crawler is an agent skill from opensanctions/opensanctions. Rewrite messy or AI-generated crawler code into clean, production-ready style that follows the zavod best practices.
When should I use Refactor Crawler?
Refactor Crawler fits situations like: the user asks to clean up; make production-ready a crawler; bring code in line with best practices.
How do I install Refactor Crawler in Claude Code?
Run `npx skills add opensanctions/opensanctions --skill refactor-crawler -a claude-code`. Or copy the skill folder (.claude/skills/refactor-crawler in opensanctions/opensanctions) into .claude/skills/refactor-crawler in your project. Claude Code loads it when a task matches its description.
How do I install Refactor Crawler in Codex?
Run `npx skills add opensanctions/opensanctions --skill refactor-crawler -a codex`. Or copy the skill folder (.claude/skills/refactor-crawler in opensanctions/opensanctions) into .agents/skills/refactor-crawler in your project. Codex loads it when a task matches its description.
Can I use Refactor Crawler 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 opensanctions/opensanctions --skill refactor-crawler -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/refactor-crawler, .gemini/skills/refactor-crawler, .github/skills/refactor-crawler and .opencode/skills/refactor-crawler in your project.
What does Refactor Crawler need to run?
Going by SKILL.md and its folder, Refactor Crawler needs the command-line tools its instructions call (ruff and mypy). Its frontmatter pre-approves these tools: Read, Edit, Write, Glob, Grep, Bash.
Does Refactor Crawler 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 Refactor Crawler safe to install?
Our automated static check of SKILL.md found notes only (pre-approves every shell command (allowed-tools: bash)), nothing it rates as a warning. It is not a guarantee. Review the folder before installing.
What licence does Refactor Crawler use?
Refactor Crawler is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Refactor Crawler use?
About 1.1k tokens (SKILL.md is roughly 4.3k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
What are the alternatives to Refactor Crawler?
Skills that share tags, products or a category with Refactor Crawler: Tmux (trpc-group/trpc-agent-go, 1.8k stars), Ketch (1broseidon/ketch, 696 stars), Crawl4AI Web Scraping (smallnest/goclaw, 598 stars) and Boss Zhipin Scraper (eatmoreduck/boss-zhipin-scraper, 1.5k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Refactor Crawler?
opensanctions (a GitHub organization) maintains it in opensanctions/opensanctions, which has 831 GitHub stars. The repository holds 11 skills in this directory. The repository was last updated on October 7, 2026.
Source: opensanctions/opensanctions on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.