agentleFS
danielrosehilldanielrosehill/Israel-Research-Skills/CLAUDE.md

Israel-Research-Skills

CLAUDE.md0 starsChanged 29 days agoMITView on GitHub

Public Claude plugin israel-research. It holds instruction-only skills for Israel-specific research and know-how (rights, civil defence, travel warnings, medicines, offices, phone numbers, opening hours, salaries, shopping). The skills must work on both Claude and ChatGPT. Created 2026-10-16. - Claude: installable directly. The repo is its own marketplace (.claude-plugin/marketplace.json, source ./). - The household ChatGPT Business workspace. The repo is vendored into danielrosehill/Residence-Skills-Marketplace-ChatGPT by danielrosehill/Residence-Skills → skills/publish-to-chatgpt/scripts/publish.sh, through its EXTERNAL map ([israel-research]="danielrosehill/Israel-Research-Skills"). - The script shallow-clones the pushed default branch. -…

The file

We split this CLAUDE.md into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  • could be automatic: a setting could enforce it
  1. Distinctive:

    Public repo: No personal identifiers, household addresses or credentials.

    line 43

≈ “Common” is an estimate: at least 1 in 5 of the 26,980 CLAUDE.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
CLAUDE.md

Israel-Research-Skills: agent notes

Public Claude plugin israel-research. It holds instruction-only skills for Israel-specific research and know-how (rights, civil defence, travel warnings, medicines, offices, phone numbers, opening hours, salaries, shopping). The skills must work on both Claude and ChatGPT. Created 2026-10-16.

Consumers

  • Claude: installable directly. The repo is its own marketplace (.claude-plugin/marketplace.json, source ./).
  • The household ChatGPT Business workspace. The repo is vendored into danielrosehill/Residence-Skills-Marketplace-ChatGPT by danielrosehill/Residence-Skills → skills/publish-to-chatgpt/scripts/publish.sh, through its EXTERNAL map ([israel-research]="danielrosehill/Israel-Research-Skills").

    • The script shallow-clones the pushed default branch.
    • It copies only .claude-plugin/, skills/ and README.md. This repo's marketplace.json is dropped in the copy.
    • It upserts the plugin entry in the ChatGPT marketplace manifest.

    So: push here first, then run publish.sh from Residence-Skills. Do not edit the vendored copy or publish.sh from this repo.

Invariants

  • Instruction-only. No .mcp.json; ChatGPT would mark the plugin Desktop only. No scripts that a skill depends on. The skills assume only a web search tool and a web fetch or browsing tool.
  • ChatGPT cannot reach the maintainer's machine. It cannot run local Python, Playwright or Chrome, or call a private MCP gateway. Any skill whose core needs those does not belong here.
  • Framework-agnostic, per the Agent Skills spec.
    • Frontmatter name equals the folder name.
    • description says what the skill does and when to use it.
    • Refer to capabilities generically ("your web search tool"). Never use vendor tool names (mcp__…, claude-in-chrome, Playwright, Tavily).
  • Each skill is self-contained. Bulky data goes in that skill's own references/, because only skills/ is vendored. Cross-references to sibling skills are by name only, as optional hand-offs.
  • Date every time-sensitive fact, with an absolute date. Say whether it was verified or inferred. That includes VAT, thresholds, fetchability, endpoints and the snapshot date. Re-verify when editing.
  • Public repo. No personal identifiers, household addresses or credentials.
  • Out of scope:
    • driving a browser or running local code;
    • bookings and submissions (Israel Post appointments, 106 reports, Netek cancellations);
    • RTL and Hebrew formatting (danielrosehill/RTL-Conformity-Skills);
    • the Jewish calendar (its own plugin);
    • things assistants do natively (news, translation).
  • Bump version in both .claude-plugin/plugin.json and .claude-plugin/marketplace.json when skill content changes.

Verification log (2026-10-16)

  • Kol Zchut. The whole site is behind Cloudflare.

    • A browser User-Agent gets 200 on /he/<article>, /w/he/api.php (search, opensearch, parse), /w/he/rest.php/v1/search/* and index.php?action=raw.
    • curl's default UA, the bare Mozilla/5.0 UA, and a cloud AI fetch tool get 403 "Just a moment…", even from an Israeli IP.
    • robots.txt disallows /w/*/api.php and /w/*/index.php for crawlers.
    • The English edition was removed 2023-02-02; /en/* redirects to a notice.
    • The Arabic and Russian APIs answer.

    Details: skills/kol-zchut-rights/references/api.md.

  • Import tax. VAT is 18%. The personal-import threshold is $75 goods value from 2026-06-02, after $150 and $130 periods. There is a 72-hour rule for same-supplier split parcels. Source: skills/personal-import-tax/references/israel-import-tax.md.
  • gov.il and taxes.gov.il: 403 Cloudflare. taxes.gov.il now redirects to gov.il.
  • Readable: btl.gov.il, iec.co.il, taasuka.gov.il, zap.co.il search, and Hebcal's Shabbat API.
  • drug.co.il: 503 to automated fetches.
  • Candle lighting. Hebcal by GeoNames ID gave Jerusalem 17:26, Haifa 17:35 and Tel Aviv 17:47, against a Jerusalem sunset of 18:06. This confirms the 40, 30 and 20 minute customs.

Sources the skills were derived from

  • danielrosehill/Claude-Israel-Agent-Skills-Plugin
  • danielrosehill/Claude-Israel-Shopping-Plugin
  • local ai-claude-plugins/:
    • Aliexpress-Israel-Skills
    • Israel-Phonebook-Manager
    • Claude-Israel-Opening-Hours-Plugin
  • israel/Israel-Telephone-Numbering
  • israel/Israel-ADHD-Medication-Access (the general rules only, not the case)

The README maps each skill to its source.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.