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 levan-ildani/georgian-grammar-codex-skill --skill georgian-grammargit clone --depth 1 https://github.com/levan-ildani/georgian-grammar-codex-skillWrote 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/levan-ildani/georgian-grammar-codex-skill/georgian-grammar)<a href="https://agentmods.dev/skills/levan-ildani/georgian-grammar-codex-skill/georgian-grammar"><img src="https://agentmods.dev/badge/skills/levan-ildani/georgian-grammar-codex-skill/georgian-grammar/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/levan-ildani/georgian-grammar-codex-skill/georgian-grammar"><img src="https://agentmods.dev/badge/skills/levan-ildani/georgian-grammar-codex-skill/georgian-grammar.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 | $0.00088 | $0.02157 |
| Opus 5 | $0.00044 | $0.01078 |
| Sonnet 5 | $0.00018 | $0.00431 |
| Haiku 4.5 | $0.00009 | $0.00216 |
Grade A, and why
georgian-grammar 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 11d 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 — 110 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Georgian Grammar
Overview
Write clean, natural, standard Georgian and reduce the typical mistakes AI agents make in morphology, syntax, agreement, politeness, conjunction choice, and punctuation. Default to contemporary literary Georgian in Mkhedruli unless the user explicitly asks for transliteration, dialect, or archaic style.
Use this skill as a working grammar filter inside Codex, not as a test-answer memory. Every Georgian answer, edit, translation, explanation, UI string, or selected multiple-choice option must pass rule-first Georgian QA before finalizing.
Workflow
- Decide whether the task is drafting, translation, editing, or QA.
- For any form-judgment task, classify the linguistic layer before answering: orthography, morphology, syntax, lexical standardness, register, or punctuation.
- Use
references/normative-decision-protocol.mdas the rule-first route, especially for short grammar questions. - Classify the risk before opening references:
- verb form or verb agreement:
references/verb-morphology.md - complex morphology, voice/version, causative, participle, or multi-step verb analysis:
references/morphology-deep-dive.md - concrete verb stem, theme sign, participle, or paradigm standardness:
references/verb-paradigm-boundary.md - joined, separate, or hyphenated writing:
references/compound-writing-rules.md - particles, fixed expressions, conjunctions, conditionals, reported-speech suffixes, or prohibitions:
references/particle-expression-rules.md - adverbs, adverbial meaning, question/relative/indefinite/negative adverbs, or adverb formation:
references/adverb-rules.md - complex syntax, indirect speech, relative clauses, word order, or multi-clause punctuation:
references/syntax-deep-dive.md - cases, postpositions, possessive/reflexive forms, demonstratives, or direction forms:
references/case-and-postposition-rules.md - subject-predicate, numeral, relative-pronoun, or
თქვენagreement:references/agreement-rules.md - spelling, fixed orthography, numerals, high-risk fixed forms, or compact spelling reminders:
references/normative-orthography.md - specific lexical paradigm, word meaning, or normative dictionary question:
references/lexical-norm-boundary.md - stylistic correctness, semantic precision, concise Georgian wording, or near-synonym choice:
references/style-and-lexical-choice.md - punctuation, direct address, or register:
references/polite-forms-and-punctuation.md
- verb form or verb agreement:
- Read
references/grammar-guardrails.mdfor the compact global checklist when drafting or revising more than a short answer. - Draft, correct, or answer from the rule. Prefer standard contemporary literary Georgian unless the user asks for colloquial, dialectal, or archaic style. If the exact rule is not in the local references, choose a safer construction or state uncertainty instead of guessing from surface naturalness.
- Do a final human-style pass with this checklist:
- verb form derived from the correct paradigm, not only from surface analogy
- subject-predicate number agreement
- numeral plus noun behavior
- correct case before postpositions
- possessive pronoun form before dative or adverbial nouns
- pronoun reference, substitution, and ambiguity
თავისიvsმისიownership- relative pronoun number and case
თქვენagreement and polite register- conjunction choice for cause, purpose, and condition
- adverbial relation: place, time, manner, cause, purpose, frequency, or negation
- prohibition forms with
არvsნუ - joined, separate, and hyphenated writing, especially lexicalized compounds
- punctuation and spacing
- natural Georgian phrasing rather than literal calques
What ships with it
20 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.
- agents/openai.yaml 322 B
- assets/georgia-flag.png 3.0 KB
- assets/georgia-flag.svg 668 B
- references/adverb-rules.md 4.9 KB
- references/agreement-rules.md 3.1 KB
- references/case-and-postposition-rules.md 7.2 KB
- references/compound-writing-rules.md 3.8 KB
- references/grammar-guardrails.md 11 KB
- references/lexical-norm-boundary.md 5.3 KB
- references/maintenance-regression-prompts.md 4.5 KB
- references/morphology-deep-dive.md 7.2 KB
- references/normative-decision-protocol.md 3.8 KB
- references/normative-orthography.md 2.2 KB
- references/particle-expression-rules.md 5.3 KB
- references/polite-forms-and-punctuation.md 6.9 KB
- references/source-map.md 3.6 KB
- references/style-and-lexical-choice.md 10 KB
- references/syntax-deep-dive.md 6.4 KB
- references/verb-morphology.md 7.5 KB
- references/verb-paradigm-boundary.md 3.9 KB
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.
- 11d ago First seen · 110 lines · 88 tokens per session scan A 5447ae351c41
georgian-grammar is a skill published in the GitHub repository levan-ildani/georgian-grammar-codex-skill (7 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 88 tokens to every session and 2,157 once invoked, about $0.0004 per session on Opus 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 2026-08-31.
Other skills, from other repositories
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.
aws-wechat-article-formatting
A local converter that turns Markdown text into HTML formatted for the Chinese WeChat editor. Markdown is plain text with simple symbols for headings, links, and emphasis; the converter puts the resulting styles directly into the HTML.
qu-ai-wei
A tool for rewriting Simplified Chinese so it sounds more natural while keeping the original facts, meaning, evidence strength, tone, and authorial voice.
speak-human-tw
A Traditional Chinese editing guide for making public-facing writing sound more natural and less machine-generated. It also checks wording associated with Mainland China and half-width punctuation.
inkos
Story Creation and Translation AI Agent with Studio Chat, CLI, and TUI - use for long-form novels, short fiction, scripts, storyboards, interactive-film projects, open-world / branching play, fan fiction, spinoffs, style imitation, continuations, covers, and multilingual EPUB/PDF/TXT/Markdown translation. Includes…
humanizer-ru
A Russian-text editor that removes bureaucratic wording, filler, copied chatbot patterns and signs of AI writing while keeping facts and examples.