Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add majdghithan/agent-skills --skill lahjagit clone --depth 1 https://github.com/majdghithan/agent-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/majdghithan/agent-skills/lahja)<a href="https://agentmods.dev/skills/majdghithan/agent-skills/lahja"><img src="https://agentmods.dev/badge/skills/majdghithan/agent-skills/lahja/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/majdghithan/agent-skills/lahja"><img src="https://agentmods.dev/badge/skills/majdghithan/agent-skills/lahja.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00307 | $0.03164 |
| Opus 5.5 | $0.00123 | $0.01266 |
| Sonnet 5.5 | $0.00061 | $0.00633 |
| Haiku 4.5 | $0.00031 | $0.00316 |
Grade A, and why
lahja scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 25d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 128 lines — stays where its author put it; the contents beside it link to each section on GitHub.
lahja - write Arabic in a real dialect
Built on the fasih method. lahja follows the approach of fasih by Maher El Gamil: write Arabic that sounds native rather than translated, adapt instead of rendering word for word, never invent a fact or a form, and keep the register consistent. fasih covers Arabic levels and tone and deliberately keeps dialect short; lahja is the dialect-grammar layer under that same method.
AI dialect text fails on grammar, not vocabulary: it swaps أريد for بدي but keeps سوف, لقد, أن + verb, and MSA negation, so it still reads as translated فصحى. This skill fixes the grammar layer. Pick one dialect, load its guide, write through it.
Step 1 - Detect the dialect FIRST (ask at most one question)
In order, stop at the first that answers:
- Named in the request: a dialect, country, or city ("فلسطيني", "Egyptian", "for a Kuwaiti audience").
- The user's own Arabic is clearly in one dialect -> match it. "Clearly" = at least two markers from one row below and none from another row. Forms shared across dialects don't count, however dialect-y they feel. The verified overlaps:
Form Shared by رح + verb Levantine and Iraqi شلون Gulf and Iraqi هاي Levantine and Iraqi قاعد + verb Gulf, Iraqi, rural Levantine, Egyptian, Tunisian شنو Gulf, Iraqi, Moroccan, Tunisian أكو / ماكو Iraqi and Kuwaiti فين Egyptian, Moroccan, Tunisian وين Levantine, Gulf, Iraqi, Algerian, Tunisian إيش / آش Levantine, Hejazi, Moroccan, Tunisian كمان Levantine and Egyptian شو Levantine and Emirati ده / دي Egyptian and Sudanese دا Iraqi (progressive) and Sudanese (demonstrative) أبي / يبي Gulf, Libyan, Yafi'i Yemeni ماش / ماشي Yemeni (negator) and Maghrebi (negator or future) قداش Maghrebi and Libyan غدوة Yemeni and Libyan مش، بس، اللي، ما في، كتير، عشان، ليش، اكتبلي most dialects
What ships with it
10 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- references/egyptian.md 11 KB
- references/gulf.md 12 KB
- references/iraqi.md 13 KB
- references/levantine.md 16 KB
- references/libyan.md 13 KB
- references/maghrebi.md 16 KB
- references/sudanese.md 12 KB
- references/yemeni.md 13 KB
- scripts/check_output.py 3.7 KB runs code
- scripts/check_skill.py 4.1 KB runs code
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 25d ago First seen · 128 lines · 307 tokens per session scan A d71268502c1f
lahja is a skill published in the GitHub repository majdghithan/agent-skills (8 stars, last pushed 25d ago), licensed MIT. It adds 307 tokens to every session and 3,164 once invoked, about $0.0012 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-21.
Other skills, from other repositories
olakunlevpn-filament-skills
Use when writing, reviewing, or refactoring Filament PHP v5 code -- resources, forms, tables, infolists, actions, widgets, panels, relation managers, multi-tenancy, testing, or plugin development. Always apply these standards to all Filament work.
baoyu-translate
A translation workflow for converting articles and other text into another language. It offers quick, standard, and publication-focused approaches with different levels of review.
create-powergrid-plugin
Create a complete PowerGrid plugin from scratch, including PHP class, Column macro, Blade view, Alpine.js component, and registration.
extract-source-sample
Given the path to a finished content-goose ad-run folder, extract everything that defines that ad — recipe shot list, VO script, characters, voices, world, atom-skills, master mp4 — and emit a source-sample.json in the exact shape the upload-ad-sample skill writes to the Goose Ads library. Also links every character…
hreflang-international
International / multilingual SEO audit focused on hreflang correctness: missing or broken return tags, wrong language/region codes, missing x-default, self-referencing errors, hreflang vs canonical conflicts, and inconsistent signals across HTML head, HTTP headers and XML sitemap. Also reviews URL structure (ccTLD vs…
i18n-parity
MANDATORY for every coding agent and contributor touching localized content — keep all five localization surfaces (dashboard UI keys, wiki page, full mirrored guides, locale-aware formatting, language switchers) in parity across every supported language. Use automatically (without being asked) whenever you add or…