agentleFS
danielrosehilldanielrosehill/Israel-Research-Skills/skills/kol-zchut-rights/SKILL.md

kol-zchut-rights

Skill0 starsChanged 31 days agoMITView on GitHub

Research Israeli rights and entitlements on Kol Zchut (כל זכות, kolzchut.org.il), the plain-language rights wiki, and answer with cited article links. Use for any question of the form "what are my rights when…", "am I entitled to…", "what does Israeli law say about…" — employment (severance, notice, sick days, recuperation pay, overtime), National Insurance (Bituach Leumi) benefits, maternity and parental leave, reserve duty, tenancy, consumer cancellations, health basket, tax credits, olim and returning residents, disability, elderly, debts and enforcement. Covers Hebrew search terms by domain, the public MediaWiki API, how to cite, and when to escalate to gov.il or Bituach Leumi.

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 4 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Hebrew is the main edition and the one to use: Search in Hebrew even when the user writes in English, then answer in the user's language.

    line 16
  2. Very specific:

    Translate the question into Hebrew terms: Use the domain table in references/hebrew-search-terms.md.

    line 31
  3. Distinctive:

    Read the article before answering: Do not answer substance from a snippet.

    line 44
  4. Distinctive:

    Do not loop on a refused URL.

    line 135
  5. Very specific:

    Never invent an article title or URL.

    line 139
  6. Distinctive:

    Keep Hebrew legal terms next to their English gloss, for example "recuperation pay (דמי הבראה)".

    line 142

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill 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
SKILL.md

Kol Zchut rights research

Kol Zchut (כל זכות, "All Rights") at https://www.kolzchut.org.il is a nonprofit wiki that explains Israeli rights and entitlements in plain language. For any question that hinges on Israeli law or benefits, it is the first source to check, before general web results or memory. Each article states who is eligible, how to claim, and links to the law and the responsible authority.

Know what you are searching

  • Hebrew is the main edition and the one to use. Search in Hebrew even when the user writes in English, then answer in the user's language.
  • There is no English edition. It was removed on 2 February 2023, and kolzchut.org.il/en/... URLs now redirect to a Hebrew notice saying so (verified 2026-10-16). Do not send English speakers to /en/, and do not cite an old English page from a search cache as current.
  • Arabic (/ar/) and Russian (/ru/) editions exist and have their own APIs. They cover fewer topics than Hebrew. Offer them to Arabic or Russian speakers, but check the Hebrew article for anything time-sensitive.
  • Not for: case law, court procedure or a specific ruling. Kol Zchut summarises rights, not litigation. For that, point the user to Nevo or Takdin (paid legal databases) or to a lawyer.

Workflow

  1. Translate the question into Hebrew terms. Use the domain table in references/hebrew-search-terms.md. Most terms there are exact article titles, verified against the site on 2026-10-16. Prefer a specific term (הודעה מוקדמת לפיטורים) to a broad one (עבודה), because broad terms return portal pages.
  2. Find the article. Try these in order and stop at the first that works:
    • Web search restricted to the site: site:kolzchut.org.il <Hebrew terms>. This works in every environment, and search snippets often already show the article title.
    • The site's search API, if your fetch tool can read it. It returns ranked titles with snippets as JSON. Patterns are in references/api.md.
    • The article URL directly, if you already know or can guess the title: https://www.kolzchut.org.il/he/<title with spaces as underscores>.
  3. Read the article before answering. Do not answer substance from a snippet. If your fetch tool is blocked (see "When fetching fails"), say that you are working from search snippets, and give the user the link.
  4. Answer from the article.
    • Lead with the direct answer: entitled or not, how much, how long.
    • Give the conditions that decide eligibility.
    • Give the steps to claim and who handles it.
    • Quote figures exactly as the article states them, including the date they apply from. Kol Zchut updates amounts such as minimum wage, benefit rates and the reserve-duty compensation floor when they change, and old figures linger elsewhere on the web.
  5. Cite. See the citing section below.

Reading an article

Articles share a structure, so you can go straight to the part you need:

Section (Hebrew)What it holds
Summary box at the top (תקציר)One-paragraph answer
מי זכאי / תנאיםEligibility conditions
כיצד מממשים את הזכותHow to claim: forms, deadlines, which office
גורמים מסייעיםWho can help: government call centres, aid organisations, the responsible authority
מקורות משפטיים ורשמיים / חקיקה ונהליםThe statute, regulations and official circulars
פסקי דיןSelected rulings
זכותון ... pagesRights digests: a whole topic on one page, for example domestic workers or personal imports. Good for overview questions.

Portal pages (פורטל) list every article in a domain. Use one when the user's situation does not match a single article. For example, the עולים page is the right starting point for a new immigrant.

Citing

Give the article title in Hebrew, with an English gloss, and its full URL. Add the last-updated date if you can see it. The footer of each article shows עודכן לאחרונה ב־<date>, and the API returns touched.

Source: Kol Zchut — "פיצויי פיטורים" (Severance pay)
https://www.kolzchut.org.il/he/פיצויי_פיטורים  (last updated <date>)

Some chat surfaces break Hebrew URLs. If that happens, use the percent-encoded form, which the search API returns as fullurl, or which you can build by URL-encoding the title. When several articles are relevant, cite each one next to the claim it supports.

When to escalate to an official source

Kol Zchut is secondary and says so: its footer states that binding rights come from statute, regulations and court rulings. Go to the primary source, and tell the user you did, when:

  • The user is about to act on a number. Benefit amounts, contribution rates and thresholds come from National Insurance (btl.gov.il), the Tax Authority (taxes.gov.il) or the relevant ministry page on gov.il. Each Kol Zchut article links to the responsible body under גורמים מסייעים.
  • The article is visibly stale, or the topic changed recently. This is common for wartime arrangements such as חרבות ברזל benefits, which are often temporary and extended by order.
  • A deadline or a claim decision matters, such as an appeal window, a medical committee (ועדה רפואית) or a dismissal hearing (שימוע).
  • The question is about one specific case (a dismissal, an eviction, a benefit refusal). Give the rights from Kol Zchut and the relevant aid organisation from גורמים מסייעים. Mention free legal aid (סיוע משפטי חינם) where it applies.

btl.gov.il usually fetches cleanly. www.gov.il blocks most automated fetches with a Cloudflare challenge. taxes.gov.il now redirects there too (checked 2026-10-16). When a page is blocked, give the user the link, or ask them to paste the page, rather than retrying.

When fetching fails

The whole site sits behind Cloudflare. As tested from an Israeli IP on 2026-10-16:

  • Requests that look like a browser got through: articles, api.php and rest.php all returned content.
  • Requests with a generic or bot User-Agent got 403 and a "Just a moment…" page. That covered plain curl and one AI fetch tool.
  • robots.txt disallows /w/*/api.php and /w/*/index.php for crawlers, but allows article paths (/he/, /ar/, /ru/).

So your fetch tool may be refused even though the site is up. If it is:

  1. Rely on site:kolzchut.org.il web search. Snippets and titles still identify the right article.
  2. Try the article URL once in a real browsing tool, if you have one.
  3. Give the user the article link, and say you could not open the full text.

Do not loop on a refused URL.

Answer hygiene

  • Never invent an article title or URL. If search finds nothing, say so, and suggest the user search the site directly: https://www.kolzchut.org.il/he/מיוחד:חיפוש?search=<terms>.
  • Keep Hebrew legal terms next to their English gloss, for example "recuperation pay (דמי הבראה)". The user will meet the Hebrew term on forms and payslips.
  • Close every answer about a specific case with exactly this line: Kol Zchut is general information, not legal advice. For a specific case, check with a lawyer or the responsible authority.

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.